1 00:00:00,634 --> 00:00:02,284 On this episode of restore it all. 2 00:00:02,284 --> 00:00:06,064 We've got a new solution for the problem of ransomware attacking 3 00:00:06,064 --> 00:00:07,564 Windows-based backup servers. 4 00:00:07,894 --> 00:00:11,074 This one's aimed specifically at Veeam, but it looks like there 5 00:00:11,074 --> 00:00:12,784 are many other applications. 6 00:00:12,964 --> 00:00:14,344 So hope you enjoy this episode. 7 00:00:38,181 --> 00:00:41,391 W. Curtis Preston: Hi, and welcome to Backup Central's Restore it all podcast. 8 00:00:41,391 --> 00:00:44,031 I'm your host, w Curtis Preston, aka a Mr. 9 00:00:44,031 --> 00:00:44,541 Backup. 10 00:00:44,841 --> 00:00:47,841 And I have with me the guy that I think is gonna help me 11 00:00:47,841 --> 00:00:49,551 find a new recording platform. 12 00:00:50,341 --> 00:00:52,551 Prasanna Malaiyandi how's it going? 13 00:00:52,551 --> 00:00:53,101 Prasanna 14 00:00:53,290 --> 00:00:54,130 Prasanna Malaiyandi: I'm good. 15 00:00:54,130 --> 00:00:54,460 Curtis. 16 00:00:54,460 --> 00:00:56,860 I, well, so I don't think it's all doom and gloom for the 17 00:00:56,860 --> 00:00:58,790 podcast recording platforms. 18 00:00:59,920 --> 00:01:02,530 Um, we'll just have to wait and see. 19 00:01:02,750 --> 00:01:05,951 W. Curtis Preston: Let me introduce today's guests . This is a unique one. 20 00:01:05,981 --> 00:01:09,191 I've known and known of our two guests today. 21 00:01:09,191 --> 00:01:10,151 And by the way, it's unique. 22 00:01:10,151 --> 00:01:12,401 We don't, we rarely have two guests. 23 00:01:12,851 --> 00:01:15,071 I'm gonna have to figure out how to fit you on the, on 24 00:01:15,076 --> 00:01:16,721 the, uh, Brady Bunch screen. 25 00:01:17,471 --> 00:01:23,951 I've known of one guest for almost as long as I've been in backups, and I 26 00:01:23,956 --> 00:01:28,631 was an admirer of his early work, and we'll talk about that a little bit. 27 00:01:29,711 --> 00:01:33,551 And then our other guest, I've known him for quite a while as well. 28 00:01:33,941 --> 00:01:38,141 And, uh, we've, we've gotten in trouble a little bit here and there, uh, together. 29 00:01:38,441 --> 00:01:43,491 So first I wanna welcome the c e O of Grau Data, Herbert Grau 30 00:01:43,726 --> 00:01:45,251 Thanks for coming on the podcast 31 00:01:46,634 --> 00:01:49,654 Herbert Grau: Thank you gentlemen for inviting. 32 00:01:50,698 --> 00:01:55,483 W. Curtis Preston: and, uh, And welcome of course to David Cerf. 33 00:01:55,693 --> 00:01:56,143 How's it going, 34 00:01:56,485 --> 00:01:57,325 David Cerf: Very good, Curtis. 35 00:01:57,325 --> 00:01:57,505 Good. 36 00:01:57,505 --> 00:01:57,895 See you. 37 00:02:00,363 --> 00:02:04,053 W. Curtis Preston: So I, so by the way, the fact, you know, I knew, I knew that 38 00:02:04,053 --> 00:02:08,193 we were talking to GR data today and I knew that, or at least I believed 39 00:02:08,193 --> 00:02:13,833 at the time that you were, uh, the same company or a follow on company 40 00:02:13,833 --> 00:02:18,933 from the company that I knew way back in the day, what I did not expect. 41 00:02:19,623 --> 00:02:23,733 Is to have a guest whose name matched the name of the company. 42 00:02:23,733 --> 00:02:25,353 So that was a big surprise to me. 43 00:02:25,983 --> 00:02:28,713 So let's go, Herbert, let's go back in the day. 44 00:02:28,713 --> 00:02:34,233 The first time I remember seeing you or seeing, you know, hearing 45 00:02:34,233 --> 00:02:39,753 of your, your, your company were these gigantic tape libraries. 46 00:02:39,753 --> 00:02:45,963 And I remember back in the day looking at them going, that looks amazing. 47 00:02:45,963 --> 00:02:50,733 Like, it was like , they were just these ginormous tape libraries that 48 00:02:50,943 --> 00:02:55,308 here I was, I was an early Spectra Logic customer, and they had these little, 49 00:02:55,338 --> 00:02:56,958 you know, these little carousel things. 50 00:02:56,958 --> 00:03:01,848 So I was dealing with like 30 tapes and you were dealing with thousands of tapes. 51 00:03:02,088 --> 00:03:04,188 And I remember going, holy cow. 52 00:03:04,188 --> 00:03:09,048 And one, one thing I remember was that the libraries were so big and they were 53 00:03:09,048 --> 00:03:14,448 so cost effective that it actually, and, and you can correct me if I'm wrong, 54 00:03:14,448 --> 00:03:20,658 what I remember was, That it actually cost more to fill it up with tape 55 00:03:21,468 --> 00:03:23,838 than it did to buy the library itself. 56 00:03:23,843 --> 00:03:24,108 Right? 57 00:03:24,108 --> 00:03:28,158 The, the library was so large and so cost effective that that was the case. 58 00:03:28,578 --> 00:03:31,128 Um, you wanna talk a little bit about those old days, 59 00:03:31,324 --> 00:03:31,744 Herbert Grau: Yep. 60 00:03:31,844 --> 00:03:37,394 Well, I started already very early in the 1980 eighties. 61 00:03:38,384 --> 00:03:43,724 Uh, our background was machine building, so I took the company from 62 00:03:43,724 --> 00:03:47,624 my father and we were, uh, automotive suppliers and machine builders. 63 00:03:47,624 --> 00:03:53,084 So we were not a, not an IT company, and IBM brought this, uh, tape to 64 00:03:53,084 --> 00:03:55,104 the market and had no automation. 65 00:03:56,334 --> 00:04:00,149 And StorageTek was already there and IBM had nothing. 66 00:04:00,689 --> 00:04:05,909 So we filled that gap and built a tape library in the first days. 67 00:04:05,999 --> 00:04:09,749 Weird enough, without any software connected to the host. 68 00:04:10,109 --> 00:04:14,819 This came over time through the customers, but we have been building 69 00:04:14,819 --> 00:04:20,789 these really cooltape libraries and they were called mixed media libraries 70 00:04:20,789 --> 00:04:27,029 because we could automate anybody's, uh, tape drives from Hitachi, from other 71 00:04:27,029 --> 00:04:29,369 vendors as well, even in the mixed mode. 72 00:04:30,119 --> 00:04:35,489 And so we were the exact counterpart of StorageTek or from us, and we 73 00:04:35,489 --> 00:04:37,469 had kind of more an open approach. 74 00:04:38,279 --> 00:04:42,269 And, and the second generation, we introduced the Quatro Tower. 75 00:04:42,779 --> 00:04:45,419 This was a cool patent we had. 76 00:04:46,499 --> 00:04:52,469 On one Cerface, the cartridges were moving inside small towers, so we 77 00:04:52,469 --> 00:04:59,519 could have 5,000 IBM cartridges on a small footprint, and we were shipping 78 00:04:59,519 --> 00:05:01,859 tape libraries around the world. 79 00:05:02,609 --> 00:05:09,629 When eMASS joined my company and a very large library would have 30,000 IBM tapes. 80 00:05:10,379 --> 00:05:15,419 So six of these towers in a row, and a tape robot, a traveling, 81 00:05:15,419 --> 00:05:18,809 moving robot on one side and if necessary on the second side. 82 00:05:19,229 --> 00:05:24,269 So we had a double robot system and 30,000 tapes in a row, and 83 00:05:24,269 --> 00:05:25,949 what we call also could do. 84 00:05:26,969 --> 00:05:31,589 We had a special, uh, tape format implemented, called D two, which 85 00:05:31,619 --> 00:05:33,359 only the US government had. 86 00:05:34,259 --> 00:05:38,369 If you want, I still have on my, on my drawer here an a 87 00:05:38,369 --> 00:05:41,519 d two tape from these days. 88 00:05:42,029 --> 00:05:48,269 And we converted our tape libraries to this special technology and then we 89 00:05:48,274 --> 00:05:56,609 ship through eMass to the famous unknown customers, to the famous government agency 90 00:05:56,729 --> 00:06:05,819 under us, the NSA board in 1995, a tape library with a capacity of 400 terabyte. 91 00:06:06,659 --> 00:06:10,469 At that time, my biggest customer of the Deutsche Bank had. 92 00:06:12,903 --> 00:06:13,383 W. Curtis Preston: Wow. 93 00:06:13,424 --> 00:06:17,744 Herbert Grau: said, holy cow, who in the world needs 400 terabyte? 94 00:06:18,194 --> 00:06:19,304 And what for? 95 00:06:19,634 --> 00:06:24,314 But then somebody from eMass explained me what these guys were doing, all 96 00:06:24,314 --> 00:06:28,814 these satellites in the Iraq and they had eight supercomputers from 97 00:06:28,814 --> 00:06:32,264 Cray and this was awfully expensive. 98 00:06:33,614 --> 00:06:40,904 So all this data I came from satellite in, Longley or whatever down to earth 99 00:06:40,964 --> 00:06:47,264 in this, uh, data center underground and eight super commuters and then an HS 100 00:06:47,354 --> 00:06:49,934 m of the early days called file serve. 101 00:06:51,254 --> 00:06:54,434 I think, I think Quantum is still selling this today. 102 00:06:56,444 --> 00:06:58,094 and files serve moving data. 103 00:06:58,364 --> 00:06:58,694 Yeah. 104 00:06:58,694 --> 00:07:01,334 Files serve and moving data to tape out. 105 00:07:01,339 --> 00:07:06,434 And this was of course a breakthrough from a small company because then we 106 00:07:06,434 --> 00:07:10,544 sold big time, uh, machines to the D O D. 107 00:07:11,324 --> 00:07:14,564 And in, in the US of course, unfortunately. 108 00:07:14,564 --> 00:07:19,574 Then, uh, eMAss, uh, actually the mother company of eMAss is 109 00:07:19,574 --> 00:07:21,404 Systems got bought by Raytheon. 110 00:07:22,154 --> 00:07:28,874 So the missile biased, the satellite, and then the whole thing got difficult 111 00:07:28,874 --> 00:07:32,024 and they wanted to sell this off, and I couldn't buy my company back. 112 00:07:32,024 --> 00:07:39,974 So I sold my remaining shares then to ADIC bought then eMass and I had restarted 113 00:07:39,974 --> 00:07:45,974 Crau data in Germany, again, sold my shares, and two weeks later I was on the 114 00:07:45,974 --> 00:07:51,944 market again with a new company, Crau Data, which is the company today because 115 00:07:52,094 --> 00:07:58,124 they name was not so important anymore because eMASS wanted to have eMAss data 116 00:07:58,124 --> 00:08:00,464 storage, and I was Crau data storage. 117 00:08:01,634 --> 00:08:05,984 Uh, well, Eddie wanted to get me back, but then I said, no, I, I do it on my own. 118 00:08:06,989 --> 00:08:10,269 And then we entered the market first, again, with atape 119 00:08:10,289 --> 00:08:11,579 library called Infini Store. 120 00:08:11,579 --> 00:08:15,779 But this was already an appliance software, server, 121 00:08:15,959 --> 00:08:18,119 disk, and tape in one device. 122 00:08:19,049 --> 00:08:25,829 And we sold this nicely in Germany until, uh, one point in time. 123 00:08:26,129 --> 00:08:30,039 It was not possible anymore for a small company to sell hardware. 124 00:08:31,709 --> 00:08:35,399 And then we had extremely nice products, hardware, products,tape libraries, new 125 00:08:35,404 --> 00:08:40,829 generations, smaller, uh, easy, lean, cost effective, but we had to sell this. 126 00:08:41,489 --> 00:08:46,799 And then I met David, this was about 2 0 7, and then I restarted the 127 00:08:46,799 --> 00:08:50,309 company again, the same company, but they restarted as a software company. 128 00:08:51,203 --> 00:08:51,813 W. Curtis Preston: Mm-hmm. 129 00:08:52,139 --> 00:08:56,189 Herbert Grau: So, and then of course, tape, H S M was our background. 130 00:08:56,194 --> 00:09:02,879 So we had a product, which in ib, um, H P E O, emd, it was on their 131 00:09:02,879 --> 00:09:05,289 price list as file system extender. 132 00:09:06,549 --> 00:09:06,709 W. Curtis Preston: Mm-hmm. 133 00:09:06,709 --> 00:09:07,129 But not 134 00:09:07,129 --> 00:09:10,289 Herbert Grau: very successful because HPE was in the terminal, 135 00:09:10,289 --> 00:09:14,879 so many products, and we kind of were sitting between the chairs. 136 00:09:15,539 --> 00:09:20,484 And then HPE stopped the contract and we sold it under the Grau logo. 137 00:09:20,484 --> 00:09:23,604 And over time we worked our portfolio. 138 00:09:23,604 --> 00:09:28,974 And in the last four years, we have developed a complete, almost 139 00:09:29,374 --> 00:09:33,594 complete new product portfolio, which now re uh, looks really good. 140 00:09:34,044 --> 00:09:38,034 And that's why I'm, I'm in the mode of re-entering to the US 141 00:09:38,034 --> 00:09:42,474 with my friend David, and sell our nice products to the us. 142 00:09:42,474 --> 00:09:46,254 And I have been coming and traveling all along the last years. 143 00:09:47,244 --> 00:09:52,314 Still have partners and friends and, and no customers anymore, but 144 00:09:52,314 --> 00:09:54,534 this will hopefully change soon. 145 00:09:56,173 --> 00:09:56,663 W. Curtis Preston: Yeah. 146 00:09:56,668 --> 00:10:01,093 You know, it's interesting, David, you know, when, when we started talking, um, 147 00:10:01,603 --> 00:10:05,773 about, you know, I, I discovered this other product, this newer product, right? 148 00:10:06,483 --> 00:10:08,343 And I had no idea. 149 00:10:09,433 --> 00:10:13,933 Right, because I, I think this latest product is absolutely going after a 150 00:10:13,933 --> 00:10:17,833 problem that is really important, right? 151 00:10:18,133 --> 00:10:22,393 Um, the, the, the slight, the slight problem called ransomware, right? 152 00:10:22,813 --> 00:10:31,873 Um, and, um, in fact, I just, just a week ago I came out with an article 153 00:10:31,873 --> 00:10:36,253 in Network World that talks the, the title was, uh, ransomware is 154 00:10:36,253 --> 00:10:37,633 Coming for Your Backups, right? 155 00:10:37,633 --> 00:10:39,913 That's coming for your backup server specifically. 156 00:10:40,423 --> 00:10:45,853 And the, this latest product is, is, is aiming at solving that 157 00:10:45,853 --> 00:10:49,423 really new, challenging problem. 158 00:10:49,933 --> 00:10:56,773 Um, but, uh, I had no idea when, when we started talking that we were gonna be 159 00:10:56,773 --> 00:11:02,508 talking about a person that, that I've , that I've been involved with for 30 years. 160 00:11:02,748 --> 00:11:04,398 David, what, what do you think? 161 00:11:04,403 --> 00:11:10,638 Um, Uh, what's your goal as you, as you move this company in, you know, 162 00:11:10,638 --> 00:11:13,068 into the us or to expand it into the us 163 00:11:13,965 --> 00:11:16,155 David Cerf: Well, Curtis' phase we're at his awareness. 164 00:11:16,275 --> 00:11:16,935 First of all. 165 00:11:16,995 --> 00:11:21,105 Um, I think you're, it was funny how we did reloop together, which was 166 00:11:21,105 --> 00:11:23,895 that Dave Russell and I were talking and he had mentioned he had heard 167 00:11:23,895 --> 00:11:26,565 this podcast talking about Blocky for 168 00:11:26,565 --> 00:11:27,415 Veeam 169 00:11:27,435 --> 00:11:29,905 which is the product you were just mentioning, and he 170 00:11:29,955 --> 00:11:31,125 didn't mention it was you. 171 00:11:31,695 --> 00:11:34,665 And so I had to go look it up, and then I was like, well, it's Curtis. 172 00:11:34,695 --> 00:11:35,355 Uh, wow. 173 00:11:35,355 --> 00:11:36,585 And I, so we reached out. 174 00:11:36,585 --> 00:11:40,825 And so, uh, so there's, the point is that awareness issue is that, um, Grau 175 00:11:40,845 --> 00:11:43,575 has done exceptionally well in Europe. 176 00:11:44,190 --> 00:11:48,190 Uh, working with channel partners and around, uh, the, the customer 177 00:11:48,195 --> 00:11:51,960 base, uh, for, because as you mentioned, uh, ransomware is 178 00:11:51,965 --> 00:11:53,580 such a critical issue right now. 179 00:11:54,120 --> 00:12:00,060 And, um, the, the way the blocky product works is a zero trust that in it 180 00:12:00,060 --> 00:12:04,950 really, uh, brings a level of security to the large, uh, the largest install 181 00:12:04,950 --> 00:12:06,900 base for Veeam or these Windows users. 182 00:12:07,260 --> 00:12:09,420 And, uh, this gives 'em a very simple, easy. 183 00:12:10,365 --> 00:12:15,435 Quick solution, and that went like wildfire through the reseller partners go. 184 00:12:15,645 --> 00:12:17,325 So Europe behaves a little different, right? 185 00:12:17,325 --> 00:12:22,245 Channels a operate a little differently than American channels and, uh, resellers. 186 00:12:22,635 --> 00:12:25,725 And so they've done incredibly well with this traction and awareness. 187 00:12:25,730 --> 00:12:30,435 So we'd like to bring that awareness and that success, uh, out of Europe, 188 00:12:30,615 --> 00:12:34,605 uh, and not just to North America, but globally because, you know, v of course 189 00:12:34,605 --> 00:12:38,625 is global, uh, has a really strong footprint in South America and Asia, 190 00:12:39,045 --> 00:12:42,225 and you have a tr and the majority of their customers are Windows users. 191 00:12:42,225 --> 00:12:45,705 So getting that message out would certainly be the, the goal. 192 00:12:45,705 --> 00:12:49,695 And I think the product speaks for itself because there are no real options. 193 00:12:49,700 --> 00:12:53,655 It's either you, either I have Windows and I do something or I don't. 194 00:12:53,745 --> 00:12:57,135 And we're that something you can do to bring security and cyber? 195 00:12:57,324 --> 00:13:00,394 Herbert Grau: Currently we have four products, three of 196 00:13:00,394 --> 00:13:02,124 them, brand, almost brand. 197 00:13:03,514 --> 00:13:07,954 And we have a product which we sell since many years very successfully. 198 00:13:07,954 --> 00:13:09,244 It's called File Lock. 199 00:13:10,114 --> 00:13:13,744 It's a Windows based software for compliant archiving. 200 00:13:14,824 --> 00:13:21,484 And we have a KPMG certificate that nobody can alter data after it has been archived. 201 00:13:22,324 --> 00:13:28,144 And we have sold this product about 1,500 times in Europe. 202 00:13:29,134 --> 00:13:32,044 And it's based on the filter driver technology. 203 00:13:32,704 --> 00:13:34,354 And it's embedded in Windows. 204 00:13:34,444 --> 00:13:39,034 So you can have it on the Windows server, very simple on the physical or virtual 205 00:13:39,034 --> 00:13:41,404 machine, just install the software. 206 00:13:41,464 --> 00:13:46,354 And this filter driver, make sure nobody, not even the admin, can 207 00:13:46,359 --> 00:13:51,184 alter data, which is supposed to be archived for 10 years or whatever. 208 00:13:51,336 --> 00:13:55,686 It has the same API as the Snap Lock API from NetApp. 209 00:13:57,445 --> 00:13:59,095 W. Curtis Preston: Oh, that's one that Prasanna should know. 210 00:13:59,413 --> 00:14:01,663 Herbert Grau: And, and the snap Lock was the role model. 211 00:14:01,668 --> 00:14:04,243 And this API is, um, not protected. 212 00:14:04,243 --> 00:14:08,983 So we have the same api, single file retention, like uh, snap 213 00:14:08,983 --> 00:14:11,113 lock but we are independent. 214 00:14:11,173 --> 00:14:15,583 We run on the Windows server and we scale as much window scales 215 00:14:15,588 --> 00:14:20,443 from hundred gigabytes for a small company to multiple terabyte in large 216 00:14:20,443 --> 00:14:22,723 sites, cluster ready, everything. 217 00:14:23,353 --> 00:14:28,213 And this product is pretty cool and stable because filter driver technology 218 00:14:28,723 --> 00:14:30,723 was not so stable 20 years ago. 219 00:14:31,933 --> 00:14:34,663 We produced blue screen in the very early days. 220 00:14:34,723 --> 00:14:36,133 All the Veeam guys asked me that. 221 00:14:36,403 --> 00:14:40,813 But since, uh, Microsoft introduced a mini filtered technology, so 222 00:14:40,813 --> 00:14:45,223 an official interface for filters more than 10 years, we have zero, 223 00:14:45,253 --> 00:14:46,753 zero problems with the product. 224 00:14:47,143 --> 00:14:48,673 Very cool product, very lean. 225 00:14:49,543 --> 00:14:55,573 And one customer said, Herbert data in file lock cannot be altered by 226 00:14:55,573 --> 00:14:57,553 nobody, not even by ransomware. 227 00:14:58,063 --> 00:15:05,923 That's cool, but I cannot buy a compliant archive for my data in my, the 228 00:15:05,923 --> 00:15:08,203 backup should be able to override it. 229 00:15:09,133 --> 00:15:13,763 So we took this idea and said, we create a new block, uh, product called Blocky. 230 00:15:14,563 --> 00:15:15,973 And this is like a filter driver. 231 00:15:15,973 --> 00:15:20,263 This is like a sheet metal plate, a warm, a warm shield. 232 00:15:20,863 --> 00:15:22,183 Nobody can go through it. 233 00:15:22,633 --> 00:15:24,433 And then we drill a small hole. 234 00:15:25,063 --> 00:15:30,733 And in the small hole, one guy says nobody can pass except the Veeam application. 235 00:15:31,663 --> 00:15:37,223 And if the Veeam application comes, this application always has to show a passport 236 00:15:37,228 --> 00:15:39,853 and a fingerprint, like if I enter the us. 237 00:15:40,183 --> 00:15:40,603 Okay? 238 00:15:42,133 --> 00:15:46,123 And that's why we can block everybody, even the good and the 239 00:15:46,123 --> 00:15:51,883 bad, except the one application which we whitelist, and that was blocky. 240 00:15:51,943 --> 00:15:55,903 We also have for IBM TSM customer, but this was the first one. 241 00:15:56,863 --> 00:16:02,323 And the selling was, uh, the, when a word, uh, you call this word on mouth, 242 00:16:03,593 --> 00:16:06,463 customer said, wow, it's explained. 243 00:16:07,498 --> 00:16:10,408 20 minutes, it's installed in 10 minutes. 244 00:16:10,708 --> 00:16:15,988 It's so effective, costs less, and is really cool and effective. 245 00:16:16,648 --> 00:16:23,368 And that's why we sold 500 customers only in German speaking countries in the last 246 00:16:23,373 --> 00:16:31,378 four years, among them pretty big names from small Soho customers to really, 247 00:16:31,438 --> 00:16:33,808 really large international corporations. 248 00:16:34,978 --> 00:16:39,208 And that was really, uh, really a home run for us because we could use 249 00:16:39,208 --> 00:16:45,238 the technology which was proven over many, many years to a different field. 250 00:16:48,137 --> 00:16:51,707 W. Curtis Preston: And, and so it sounds like this, this grew out of that, the, 251 00:16:51,737 --> 00:16:56,207 the audit proof archiving, uh, line that you had the file lock from there. 252 00:16:56,687 --> 00:16:59,927 Um, and then you also, you've also got a couple of other products. 253 00:16:59,927 --> 00:17:02,777 You talk about metadata mining and the tape object archive. 254 00:17:02,777 --> 00:17:04,117 Do you wanna talk about that a little bit? 255 00:17:04,768 --> 00:17:08,008 Herbert Grau: All of course tape is our background as we talked 256 00:17:08,728 --> 00:17:10,768 and Ta tape will never go away. 257 00:17:11,368 --> 00:17:15,388 So we have a product which we have on the market since, um, 258 00:17:15,928 --> 00:17:18,628 almost 20 years now is stable. 259 00:17:18,628 --> 00:17:22,138 Product was a bit aged classical tape hsm. 260 00:17:22,573 --> 00:17:23,683 Like other products. 261 00:17:24,253 --> 00:17:28,813 And we have customers, big customers like Max Plank Institute with 262 00:17:29,263 --> 00:17:31,033 multiple petabyte and 10 of these. 263 00:17:31,753 --> 00:17:33,613 And we have a legacy installed base. 264 00:17:34,123 --> 00:17:39,463 And some time ago we decided that we do a new architecture because 265 00:17:39,463 --> 00:17:41,863 we think tape will never go away. 266 00:17:41,893 --> 00:17:46,783 Next, whatever is, it's a niche market, but we are an expert 267 00:17:46,783 --> 00:17:48,283 in this niche, niche market. 268 00:17:48,733 --> 00:17:54,313 And I have customers which I want to lead to the next generation. 269 00:17:54,793 --> 00:17:58,333 And that's why we developed a product called Extreme Store. 270 00:17:59,693 --> 00:18:06,103 And this is now an object storage product is a scalable object 271 00:18:06,103 --> 00:18:08,923 storage software with S3 to tape. 272 00:18:10,423 --> 00:18:11,683 That's a difference. 273 00:18:12,283 --> 00:18:13,843 And with this object, 274 00:18:14,267 --> 00:18:17,327 W. Curtis Preston: I, I interface with it via the S3 275 00:18:17,327 --> 00:18:19,047 protocol, and then you put it on 276 00:18:19,243 --> 00:18:21,928 Herbert Grau: Uh, Maybe you know the Black Pearl from Spectrum 277 00:18:21,928 --> 00:18:23,308 Logic, because we mentioned that 278 00:18:23,308 --> 00:18:26,578 name and that's kind of a product. 279 00:18:27,208 --> 00:18:30,388 Well, we not, we compete, not so much in Germany, but in the US 280 00:18:30,393 --> 00:18:32,788 this would be our major competitor. 281 00:18:33,598 --> 00:18:38,188 But this is a market where only very few companies play. 282 00:18:38,908 --> 00:18:44,908 In Europe, I see two and we have I think, the best architecture, architecture. 283 00:18:45,898 --> 00:18:49,258 Um, we have a scalable architecture. 284 00:18:49,263 --> 00:18:51,088 We have, uh, no SQL database. 285 00:18:51,093 --> 00:18:55,258 We can scale this vertically into multi-billions and 286 00:18:55,258 --> 00:18:57,358 horizontally into multi-services. 287 00:18:57,988 --> 00:18:59,818 And important in the tape world. 288 00:19:00,058 --> 00:19:05,258 If we have very small files and you have billions, you have to do containers. 289 00:19:06,598 --> 00:19:12,958 You cannot put small files on tape and retrieve a billion files 290 00:19:12,958 --> 00:19:14,638 from tape without containers. 291 00:19:15,388 --> 00:19:18,148 And that's why this container technology is important. 292 00:19:18,148 --> 00:19:23,183 And we recently did a test in a partner data center of 1.5 293 00:19:23,183 --> 00:19:26,758 billion files in one bucket. 294 00:19:27,568 --> 00:19:29,758 And this is endless, scalable. 295 00:19:30,028 --> 00:19:31,078 That's important. 296 00:19:32,548 --> 00:19:37,048 And then of of course we have a modern, modern web ui. 297 00:19:37,468 --> 00:19:40,888 Some guys like still the command light interface, but more and 298 00:19:40,893 --> 00:19:43,108 more younger guys on the web ui. 299 00:19:44,008 --> 00:19:48,178 And so we have some cool things around the product, which is 300 00:19:48,178 --> 00:19:50,968 in this niche, a cool product. 301 00:19:51,838 --> 00:19:53,638 And now I have mentioned three. 302 00:19:54,148 --> 00:19:58,978 And the three would normally be good enough for a company Grau data with 30 303 00:19:58,978 --> 00:20:02,488 people having an archival background. 304 00:20:03,808 --> 00:20:04,318 But. 305 00:20:04,948 --> 00:20:08,278 I have a new product and that's really a cool product and 306 00:20:08,278 --> 00:20:10,018 that's called the Meta Data Hub. 307 00:20:11,278 --> 00:20:12,988 Why do I have this product? 308 00:20:13,888 --> 00:20:16,948 Because my friend David Cerf came four years ago. 309 00:20:17,548 --> 00:20:23,308 He was just leaving his beloved company, StrongBox and said, Herbert, 310 00:20:23,668 --> 00:20:25,528 you have to look at metadata. 311 00:20:26,278 --> 00:20:27,268 And I said, why? 312 00:20:27,958 --> 00:20:28,858 This is old stuff. 313 00:20:28,858 --> 00:20:34,438 Metadata is old stuff because we use metadata like everybody else 314 00:20:34,438 --> 00:20:43,258 since 20 years, file size and last access, and this is HSM of old school. 315 00:20:45,064 --> 00:20:45,364 Prasanna Malaiyandi: Yep. 316 00:20:45,508 --> 00:20:48,058 Herbert Grau: But if I explain you today that we have a very unique 317 00:20:48,058 --> 00:20:50,128 product, people say, how can that be? 318 00:20:51,538 --> 00:20:57,118 Because if you, if you Google metadata, you find so many products which mention 319 00:20:57,123 --> 00:21:03,718 this, you have to define metadata as standard file system metadata. 320 00:21:04,363 --> 00:21:05,983 Which is simple 321 00:21:07,779 --> 00:21:08,339 Prasanna Malaiyandi: as useful. 322 00:21:09,373 --> 00:21:12,553 Herbert Grau: it's, it's useful for many virus scanners and 323 00:21:12,553 --> 00:21:14,323 everybody, but it's simple. 324 00:21:15,043 --> 00:21:23,173 And then you have embedded metadata, and then you take very special file 325 00:21:23,173 --> 00:21:29,773 formats and you go to a research lab, you go to a Max blank Institute, 326 00:21:29,773 --> 00:21:34,423 which partners with Harvard, and they've won the special file format, 327 00:21:34,423 --> 00:21:36,853 which comes from the NASA nifty file. 328 00:21:37,453 --> 00:21:38,803 Who, who is, what is that? 329 00:21:39,643 --> 00:21:44,113 And then you look into this nifty file, for example, and this 330 00:21:44,113 --> 00:21:48,083 file has 10,000 metadata tags. 331 00:21:48,793 --> 00:21:49,633 Holy cow. 332 00:21:49,903 --> 00:21:50,833 10,000. 333 00:21:52,063 --> 00:21:52,833 And we. 334 00:21:53,833 --> 00:22:01,543 Developed a technology, how to extract these 10,000 embedded metadata tags 335 00:22:02,533 --> 00:22:05,113 and write them into a huge database. 336 00:22:06,043 --> 00:22:13,963 And now the research guy can say, I need all files which have this whatever 337 00:22:13,963 --> 00:22:16,003 dimension here and this dimension there. 338 00:22:16,003 --> 00:22:22,993 He does a Google kind of complex search and out of his 10 million files, which 339 00:22:22,993 --> 00:22:29,983 are somewhere, he gets the right 2000 files and he can narrow this down 340 00:22:29,983 --> 00:22:33,973 from 10,000 to 10,000 to 5,000, 2000. 341 00:22:34,513 --> 00:22:38,413 And then he has the right data and that's our job. 342 00:22:38,983 --> 00:22:45,913 Find the right data and we deliver them the right data to a CAR E platform, to 343 00:22:45,913 --> 00:22:48,733 an algorithm to improve it and whatnot. 344 00:22:49,213 --> 00:22:52,813 Because I have, although another company which is doing only. 345 00:22:53,818 --> 00:22:58,468 Medical data and we have huge amount of data, but you always need the 346 00:22:58,468 --> 00:23:02,578 right amount, the right data, and that's the job of the metadata hub. 347 00:23:03,328 --> 00:23:04,828 And then we go to the next one. 348 00:23:05,818 --> 00:23:08,698 This institute has a microscope from Chase. 349 00:23:09,058 --> 00:23:10,678 Very special file format. 350 00:23:11,248 --> 00:23:11,998 Holy cow. 351 00:23:11,998 --> 00:23:14,038 8,000 metadata tags. 352 00:23:14,488 --> 00:23:16,378 Next one, bioinformatic. 353 00:23:17,158 --> 00:23:22,678 I never heard these names before, but now we have a technology how to extract this. 354 00:23:22,678 --> 00:23:24,748 That's why I call it deep data mining. 355 00:23:24,753 --> 00:23:27,868 We drilled holes very, very deep. 356 00:23:28,288 --> 00:23:29,698 Same is an automotive. 357 00:23:31,718 --> 00:23:36,573 We have some of these here and they have a motor motor test equipment 358 00:23:37,143 --> 00:23:39,213 and this is spitting out files. 359 00:23:40,338 --> 00:23:42,648 and then we go there and they said, you know what? 360 00:23:43,218 --> 00:23:47,808 We would like to know which of these million files have the same parameter 361 00:23:47,808 --> 00:23:52,368 for minus 30 degrees, that amount of kilometer, and blah, blah, blah. 362 00:23:52,848 --> 00:23:54,798 And I said, you don't know that? 363 00:23:55,458 --> 00:23:55,758 No. 364 00:23:55,763 --> 00:23:56,028 How? 365 00:23:56,028 --> 00:24:01,458 How should we, nobody can do this manually and nobody can extract the data. 366 00:24:01,998 --> 00:24:08,118 So we build an extractor for this special file format, and that's why we are unique. 367 00:24:10,042 --> 00:24:12,892 W. Curtis Preston: You know, David, you, you, it sounds like you, 368 00:24:13,072 --> 00:24:15,112 you sort of brought up this idea. 369 00:24:15,472 --> 00:24:18,112 Did I, I'm, I'm a little bit like that. 370 00:24:18,442 --> 00:24:24,592 That last customer where, how is this not already everywhere, 371 00:24:27,108 --> 00:24:30,018 Herbert Grau: David told me, David explained me his product, 372 00:24:30,018 --> 00:24:31,158 which was a different product. 373 00:24:32,343 --> 00:24:34,773 This was all about storage management. 374 00:24:34,833 --> 00:24:38,403 All products are metadata for storage management. 375 00:24:38,403 --> 00:24:45,183 Move data around, get rid of the ice, get an in, and all this is about storage. 376 00:24:45,393 --> 00:24:49,503 And I said, I want to get out of storage. 377 00:24:49,743 --> 00:24:52,323 I don't want to sell terabytes anymore. 378 00:24:52,743 --> 00:24:55,083 I want to be in the analytics business. 379 00:24:55,143 --> 00:24:57,363 I want a Google like for metadata. 380 00:24:58,053 --> 00:25:02,703 This is a different game and we will go direction to artificial 381 00:25:02,703 --> 00:25:04,323 intelligence in the next steps. 382 00:25:04,773 --> 00:25:07,563 So we will move completely away from this. 383 00:25:08,163 --> 00:25:11,613 How many data is here and on the is on, move this back and forth. 384 00:25:11,613 --> 00:25:12,723 And this is old and cold. 385 00:25:13,713 --> 00:25:14,853 This is kind of. 386 00:25:16,134 --> 00:25:19,404 Prasanna Malaiyandi: Yeah, I, I'm just thinking about use cases other than that. 387 00:25:19,404 --> 00:25:22,074 I know the primary use cases you talked about, but just thinking 388 00:25:22,074 --> 00:25:25,314 about things like, I know Curtis, we always talk about archive, right? 389 00:25:25,314 --> 00:25:28,374 And how do you find what's been archive, because you don't know 390 00:25:28,374 --> 00:25:29,514 what server came from, right? 391 00:25:29,514 --> 00:25:32,204 You no longer have that storage perspective, right? 392 00:25:32,454 --> 00:25:36,504 And or even things like e-discovery, like use cases where it's like, Hey, 393 00:25:36,504 --> 00:25:40,914 tell me information related to this subject, or other things like that. 394 00:25:40,914 --> 00:25:45,564 It seems like what you've built, Herbert and David is sort of an ability to 395 00:25:45,564 --> 00:25:50,394 centralize all of these different file formats or unique file formats and 396 00:25:50,394 --> 00:25:54,624 provide that value to the customer so they can run these queries on their. 397 00:25:56,428 --> 00:25:57,153 Herbert Grau: absolutely. 398 00:25:57,213 --> 00:26:03,033 And I had a, well, actually David went to a Berlin research lab and kind of sold 399 00:26:03,038 --> 00:26:07,443 them the idea, but the product was not there, and I sold him the product Now. 400 00:26:07,773 --> 00:26:12,243 For a nice amount of money, and he was Mr. 401 00:26:12,243 --> 00:26:12,663 Crau. 402 00:26:12,753 --> 00:26:14,073 Finally, I have a product. 403 00:26:14,073 --> 00:26:18,153 I have been waiting three years for a product, and I said, so my guys, 404 00:26:18,333 --> 00:26:22,743 maybe we have a, a unique selling point here because this guy is searching the 405 00:26:22,743 --> 00:26:27,663 market for three years in the US and everywhere, and he didn't find at least 406 00:26:27,663 --> 00:26:29,343 one product which could do the job. 407 00:26:29,759 --> 00:26:32,869 David Cerf: So there, so there are two separate ways to look at it. 408 00:26:32,869 --> 00:26:37,159 One, one, uh, at Herbert has outlined very well, which is we're trying to 409 00:26:37,159 --> 00:26:41,329 understand how to drive our business intelligence, how, how do we, and that's 410 00:26:41,329 --> 00:26:46,949 really in the application space, which is this ability to extract that metadata to. 411 00:26:47,869 --> 00:26:51,709 Have better insights and understanding and visibility, which has really nothing 412 00:26:51,709 --> 00:26:54,169 to do with where the file may be stored. 413 00:26:54,739 --> 00:26:57,949 But what, there's a second use case, which is almost secondary, which is 414 00:26:57,949 --> 00:27:01,729 if I actually can understand what I have, then I can apply that to what 415 00:27:01,729 --> 00:27:03,889 I do with it by number of copies. 416 00:27:03,889 --> 00:27:07,039 Or does it need to have sort of compliance or where do I keep it? 417 00:27:07,039 --> 00:27:07,849 How long do I keep it? 418 00:27:08,149 --> 00:27:12,079 See that that was the origin of where I had come from was more 419 00:27:12,079 --> 00:27:14,179 in the extract that metadata. 420 00:27:14,184 --> 00:27:18,469 So the world, you could look, you know, if we, with hindsight we can say, Hey, we 421 00:27:18,469 --> 00:27:22,639 knew we had to have metadata to be able to drive the intelligence that we wanna 422 00:27:22,639 --> 00:27:24,589 drive through AI and machine learning. 423 00:27:25,039 --> 00:27:26,449 We, you can't get there without it. 424 00:27:27,409 --> 00:27:29,839 And so the, the difference would be it's the approach to it. 425 00:27:30,259 --> 00:27:33,649 And so the elegance that that's in the metadata hub 426 00:27:34,129 --> 00:27:35,959 is, is really that simplicity. 427 00:27:35,959 --> 00:27:40,039 Separate out the overhead that comes with the file management or trying 428 00:27:40,039 --> 00:27:43,339 to put a GLO, global name, space and all the other things that that. 429 00:27:43,924 --> 00:27:47,014 Herbert was referencing what I was trying to do, which was kind of all these 430 00:27:47,014 --> 00:27:52,504 various things and just focus really on the metadata and the, and so there 431 00:27:52,774 --> 00:27:57,634 are two really interesting things that were solved with us, um, which Herbert 432 00:27:57,634 --> 00:27:58,774 said, but let me just emphasize it. 433 00:27:58,834 --> 00:28:03,514 One is this rapid development capability for connecting to the file type. 434 00:28:03,704 --> 00:28:07,774 This, this was really a showstopper because if I have these unique elements 435 00:28:07,774 --> 00:28:12,394 and these customers could not connect to it, then it didn't matter what you 436 00:28:12,394 --> 00:28:13,714 would do, you had to solve that first. 437 00:28:13,714 --> 00:28:17,824 So Grau has solved that ability to a connect. 438 00:28:18,484 --> 00:28:19,894 So there, that was the first part. 439 00:28:19,894 --> 00:28:22,174 And then the second part was on the backside, which is, 440 00:28:22,264 --> 00:28:23,764 okay, I've done the extraction. 441 00:28:23,764 --> 00:28:27,064 So this is almost like, think of ETL in databases, right? 442 00:28:27,364 --> 00:28:29,014 Extract, transform and load. 443 00:28:29,014 --> 00:28:32,494 And except for with the, with the metadata hub, we're extracting, we're 444 00:28:32,494 --> 00:28:33,934 transforming, and then we're connecting. 445 00:28:34,054 --> 00:28:39,424 And so either we allow through our native user interfaces a way for the, the user 446 00:28:39,424 --> 00:28:43,354 to just be able to directly access, but more importantly, Is that we can connect 447 00:28:43,354 --> 00:28:44,704 to the tools that they're already using. 448 00:28:45,394 --> 00:28:49,434 And so this really creates this feed to where they can leverage 449 00:28:49,439 --> 00:28:54,124 that data to drive that business, accelerate what they're trying to do. 450 00:28:54,484 --> 00:28:57,004 Um, which cuz that's really what it's all about at the end of the day, right? 451 00:28:57,004 --> 00:29:00,664 They're, they have a problem to solve and we're helping them solve that. 452 00:29:00,922 --> 00:29:06,162 W. Curtis Preston: So speaking about what it's all about, let's get to the, let's 453 00:29:06,162 --> 00:29:12,552 get to the star of the show, I think here, uh, in terms of this podcast, um, 454 00:29:13,062 --> 00:29:14,712 you know, we, we, we've talked a lot. 455 00:29:14,712 --> 00:29:18,342 We've had, you know, we've had Dave on, um, you know, we've 456 00:29:18,342 --> 00:29:19,902 talked a lot about Veeam. 457 00:29:20,292 --> 00:29:25,212 We've talked a lot about just windows-based backup systems. 458 00:29:25,562 --> 00:29:28,362 Veeam being, you know, Veeam, and I think Veeam and CommVault would 459 00:29:28,362 --> 00:29:30,252 be the two biggest examples, right? 460 00:29:30,702 --> 00:29:37,272 Um, and the, the risk, I think that, That their customers are under, because 461 00:29:37,572 --> 00:29:42,702 Windows being, as we all know, the number one attack vector for ransomware, right? 462 00:29:43,152 --> 00:29:44,922 And so the worry is that. 463 00:29:45,477 --> 00:29:49,437 Uh, you know, it, it's been a while since I've installed Veeam for obvious 464 00:29:49,437 --> 00:29:53,067 reasons, but by the way, I, I, I haven't thrown out our usual disclaimer. 465 00:29:53,307 --> 00:29:54,927 This is an independent podcast. 466 00:29:54,932 --> 00:30:00,387 I work for Druva, Prasanna works for Zoom, and, uh, this is not a podcast of 467 00:30:00,387 --> 00:30:07,017 either company and the, um, the opinions that you hear are ours and, uh, also be 468 00:30:07,017 --> 00:30:11,457 sure to rate us by, uh, going to the, you know, your, your favorite podcast app. 469 00:30:11,457 --> 00:30:13,557 Give us some startups, give us some, give us some comments 470 00:30:13,694 --> 00:30:14,714 Prasanna Malaiyandi: Leave some comments. 471 00:30:14,727 --> 00:30:15,957 W. Curtis Preston: find this podcast. 472 00:30:16,377 --> 00:30:17,007 Absolutely. 473 00:30:17,012 --> 00:30:20,007 Feel free to tweet as long as Twitter is still 474 00:30:20,139 --> 00:30:20,919 David Cerf: long as it's still around. 475 00:30:21,237 --> 00:30:25,347 W. Curtis Preston: Um, and if you, if you'd like to, um, if you'd 476 00:30:25,347 --> 00:30:30,027 like to join the conversation, you can find me, uh, at WC preston on 477 00:30:30,027 --> 00:30:32,997 Twitter or w Curtis Preston at gmail. 478 00:30:33,297 --> 00:30:35,337 And, uh, we'd love to get you on the podcast. 479 00:30:35,847 --> 00:30:38,607 So, you know this concern, right? 480 00:30:38,607 --> 00:30:41,697 Specifically like the default installation. 481 00:30:42,147 --> 00:30:43,917 Is on a Windows based backup server. 482 00:30:43,917 --> 00:30:44,337 Right? 483 00:30:44,607 --> 00:30:49,467 And then, um, and, and, and even the main, even if you use Linux as another, 484 00:30:49,947 --> 00:30:55,827 uh, storage device, you, the, the main server's still on Windows, and they 485 00:30:55,827 --> 00:31:01,977 do have this, the Linux based, uh, storage device now as, as a, yeah, 486 00:31:01,977 --> 00:31:04,497 as a, as a, as an answer to this. 487 00:31:04,677 --> 00:31:07,057 David Cerf: Veeam, um, obviously with their, with their hard 488 00:31:07,057 --> 00:31:08,647 Linux server does create. 489 00:31:09,462 --> 00:31:10,842 A very robust option. 490 00:31:10,902 --> 00:31:13,902 I think the real differentiation, Curtis, is the customers. 491 00:31:14,292 --> 00:31:19,302 When you look at how many Veeam customers are, are using Lennox, when you look at 492 00:31:19,307 --> 00:31:24,132 their customer, you know, demographics, it's broken out as the majority, uh, 493 00:31:24,432 --> 00:31:28,362 the big majority or Windows users and a large part of those customers 494 00:31:28,362 --> 00:31:30,672 aren't going to put a Linux server in. 495 00:31:30,792 --> 00:31:32,982 Cause you know, the guy that's running this, he's a Windows guy 496 00:31:32,982 --> 00:31:35,832 and I'm not, you know, it's a religion thing almost at some point. 497 00:31:35,952 --> 00:31:38,502 And the larger corporations, it's outta simplicity. 498 00:31:38,742 --> 00:31:42,642 As Herbert mentioned, he's, we've got several, uh, global international 499 00:31:42,642 --> 00:31:47,142 companies and they have maybe hundred plus sites and they're not 500 00:31:47,262 --> 00:31:49,332 going to run this with this complex. 501 00:31:49,707 --> 00:31:56,067 Uh, um, deployment and where the blocky for Veeam comes in is, it's, as Herbert 502 00:31:56,067 --> 00:31:59,727 mentioned, you're talking about from, from the moment you learn about it to 503 00:31:59,732 --> 00:32:01,137 installing it is less than an hour. 504 00:32:01,142 --> 00:32:05,457 So the simplicity makes it really easy for the Windows guys that don't 505 00:32:05,462 --> 00:32:09,237 have to do anything different, and now they have a level of security to, 506 00:32:09,267 --> 00:32:12,762 for protecting that, that Windows backup volume and repository, right? 507 00:32:13,062 --> 00:32:16,752 So I think that that's really where the line of demarcation comes down 508 00:32:16,757 --> 00:32:20,142 to is if you're, if you're a data center and you're running a Linux 509 00:32:20,382 --> 00:32:23,412 environment and you're comfortable with that, you, you might go with the 510 00:32:23,412 --> 00:32:30,282 native, uh, Veeam hardened Linux, um, solution for those customers that don't. 511 00:32:30,402 --> 00:32:34,632 That's where we shine and we provide that easy, quick install that gives 512 00:32:34,637 --> 00:32:35,952 that level of protection against Fran. 513 00:32:37,182 --> 00:32:38,112 ransomware. 514 00:32:38,895 --> 00:32:42,435 W. Curtis Preston: And we've talked about that, that was one of my concerns as well. 515 00:32:42,475 --> 00:32:45,525 The, the one that you brought in, if you're, if you're an all window shop. 516 00:32:46,360 --> 00:32:51,780 I, I, I'm not sure even if the, if the Linux option is more secure than 517 00:32:51,780 --> 00:32:56,190 having another Windows box, I, I'm not sure if it is more secure because 518 00:32:56,190 --> 00:32:58,370 it's your only Linux box , right? 519 00:32:58,590 --> 00:32:58,890 Right. 520 00:32:58,890 --> 00:33:02,130 If you, if it's the only Linux box in your data center, I 521 00:33:02,130 --> 00:33:03,360 don't think that's a good idea. 522 00:33:03,360 --> 00:33:05,340 If it was your only Windows box in the data center, I 523 00:33:05,340 --> 00:33:06,360 don't think that's a good idea. 524 00:33:06,360 --> 00:33:06,690 Right? 525 00:33:07,050 --> 00:33:11,280 Uh, just having a, a separate OS that you have to maintain just for a single 526 00:33:11,280 --> 00:33:16,950 purpose, you know, uh, I've never been a fan of that, but why, why don't you, 527 00:33:17,040 --> 00:33:21,990 um, give a little bit more about how, so, you know, it, it sounds like the 528 00:33:21,990 --> 00:33:27,370 product is incredibly simple to, uh, to explain, David, uh, do you want 529 00:33:27,370 --> 00:33:29,740 to give, um, you know, an overview? 530 00:33:29,800 --> 00:33:32,230 It, it sounds like pretty easy to explain and Herbert's 531 00:33:32,230 --> 00:33:33,310 already given us an overview. 532 00:33:33,310 --> 00:33:34,240 You want to drill down a little 533 00:33:34,297 --> 00:33:35,257 David Cerf: Sure, sure. 534 00:33:35,317 --> 00:33:40,687 So, um, you know, maybe pick up where you, you, your comment about people 535 00:33:40,747 --> 00:33:42,307 adding something to their environment. 536 00:33:42,307 --> 00:33:45,187 I, I mean, I think that that's, that's the real challenge. 537 00:33:45,187 --> 00:33:51,157 Is it the, it, uh, and, and now if you add the security layer, whether it's the 538 00:33:51,297 --> 00:33:54,427 ciso, cso, whatever they're doing, as long as we're not talking about the hardened, 539 00:33:54,577 --> 00:33:56,767 uh, physical, these guys are overwhelmed. 540 00:33:56,827 --> 00:34:01,867 I mean, ransomware is, is, it's not a matter of, uh, if it will happen, 541 00:34:02,077 --> 00:34:03,727 it's a matter of when it will happen. 542 00:34:03,732 --> 00:34:08,047 I think we've reached that point and, and every, everybody else is, you know, 543 00:34:08,347 --> 00:34:11,647 confirmed that it's, um, it's going to be. 544 00:34:12,262 --> 00:34:13,732 A risk that they have to deal with. 545 00:34:13,732 --> 00:34:16,972 And so when they're looking for a solution, what we're finding is that 546 00:34:16,972 --> 00:34:20,812 the, the antivirus and all these other type of tools that are out 547 00:34:20,812 --> 00:34:25,732 there are really not able to provide, uh, a way to protect that last, your 548 00:34:25,732 --> 00:34:27,202 last resort, which is your backup. 549 00:34:27,207 --> 00:34:31,162 So when the virus gets in, uh, it's sitting there and the first 550 00:34:31,162 --> 00:34:33,262 thing they're gonna go after are those backup files, right? 551 00:34:33,262 --> 00:34:37,492 So they're gonna go disable that, attack that, and at some point later, right, 552 00:34:37,492 --> 00:34:40,732 because it's, it could be a, a Trojan horse where it's sitting there waiting 553 00:34:40,732 --> 00:34:45,802 and then it comes on, um, you know, you've got this, this problem is that they're, 554 00:34:45,862 --> 00:34:49,252 you're, you're at the mercy of whoever the attacker was and what their demand is. 555 00:34:49,732 --> 00:34:52,612 And this is where the blocky really comes in. 556 00:34:53,122 --> 00:34:57,172 Um, as Herber mentioned, what we're creating is a way to have 557 00:34:57,172 --> 00:34:58,822 cyber resiliency through zero. 558 00:34:59,662 --> 00:35:04,762 So when you enable, um, blocky, which is a simple download, so you literally, you 559 00:35:05,232 --> 00:35:09,442 download it and installs in, in less than 20 minutes, the first thing it's going to 560 00:35:09,447 --> 00:35:12,772 do is it's gonna say, what is the trust? 561 00:35:13,462 --> 00:35:15,262 We're gonna go right to creating the white list. 562 00:35:15,712 --> 00:35:19,552 And that white list is the trusted applications or process 563 00:35:19,552 --> 00:35:20,572 I should say, cuz it's Veeam. 564 00:35:20,572 --> 00:35:25,102 In this, um, in this case, and I'll, I'll leave a caveat here, is that the way, the 565 00:35:25,107 --> 00:35:29,962 way GR built, uh, blocky as a technology, it can be applied to other applications. 566 00:35:30,232 --> 00:35:34,432 We've really focused on the use case around, uh, Veeam. 567 00:35:34,437 --> 00:35:38,662 So in general, you could say I have other applications and allow other application 568 00:35:38,662 --> 00:35:43,162 access, but the way we've tuned this to the Veeam market, Veeam specific. 569 00:35:43,482 --> 00:35:46,992 And so the only processes that you're really trying to identify is what's 570 00:35:46,992 --> 00:35:52,212 going to happen from the, the Veeam process to access that repository. 571 00:35:52,602 --> 00:35:56,472 So the first thing you do is either you manually set that or we have an auto. 572 00:35:56,802 --> 00:36:00,132 You can literally turn on the auto discover and we'll, we'll 573 00:36:00,162 --> 00:36:01,332 discover those processes. 574 00:36:01,332 --> 00:36:04,122 You, it's within, you said a period, let's say 24 hours. 575 00:36:04,482 --> 00:36:07,332 You've run your backup, we know the process, you turn that off. 576 00:36:07,542 --> 00:36:09,582 And then at that moment we're at zero trust. 577 00:36:09,732 --> 00:36:13,842 And so nothing else is gonna go back in, um, from a ransomware perspective 578 00:36:13,842 --> 00:36:17,082 and alter modifier, delete, because we've now applied that worm. 579 00:36:17,502 --> 00:36:21,042 Um, and, and for those, just to clarify, write once, read many, right? 580 00:36:21,372 --> 00:36:25,542 And, um, and that nothing's gonna alter, it's immutable at this 581 00:36:25,542 --> 00:36:27,432 point, and you're now secure. 582 00:36:27,432 --> 00:36:30,092 So even if you had ransomware. 583 00:36:30,442 --> 00:36:34,012 It was already in the system at this point, they can't alter 584 00:36:34,462 --> 00:36:36,022 or, or modify those files. 585 00:36:36,022 --> 00:36:42,052 So reading the file out is simple and, uh, verifying with through the fingerprint 586 00:36:42,052 --> 00:36:46,702 where we actually capture all the related elements to that process, including the 587 00:36:46,707 --> 00:36:51,142 DLLs, and that is combined to create that unique identifying fingerprints. 588 00:36:51,142 --> 00:36:54,862 So every time there's a request to modify or write, Hey, we're 589 00:36:54,862 --> 00:36:56,632 checking, we're checking that. 590 00:36:56,632 --> 00:37:01,972 And if it's not an approved, um, trusted application, we'll alert to it. 591 00:37:01,972 --> 00:37:04,192 And so now you get two, two benefits here. 592 00:37:04,197 --> 00:37:10,282 One is you've got the security through, um, the protection of, of, uh, blocking. 593 00:37:10,282 --> 00:37:11,872 But second, now you've got some alerting. 594 00:37:11,902 --> 00:37:14,212 This is something that kind of caught me by surprise. 595 00:37:14,217 --> 00:37:16,582 When, when Herbert said, Hey, let's check this out. 596 00:37:16,852 --> 00:37:19,672 Was, uh, the first customer that I talked to is they're like, wow, I've got a. 597 00:37:21,027 --> 00:37:24,877 I could see my applications that are trying to hit that, that repository, 598 00:37:25,207 --> 00:37:29,197 and they can now get some reporting and visibility and transparency in 599 00:37:29,197 --> 00:37:30,397 what's going on in their system. 600 00:37:30,967 --> 00:37:33,427 And, uh, and they can take actions from that as well. 601 00:37:34,300 --> 00:37:34,840 W. Curtis Preston: Anything else? 602 00:37:36,010 --> 00:37:39,910 Anything outside of the already approved application would trigger an alert, I'm 603 00:37:39,937 --> 00:37:40,447 David Cerf: Correct. 604 00:37:40,627 --> 00:37:41,847 Tha thanks for clarity on that. 605 00:37:41,852 --> 00:37:42,817 That absolutely correct. 606 00:37:42,817 --> 00:37:46,057 So they can now see, hey, look, I, you know, we've had these declined items 607 00:37:46,057 --> 00:37:48,277 and so the admin now has some security. 608 00:37:48,517 --> 00:37:52,087 The second thing we did is we decoupled it so it's not tied 609 00:37:52,087 --> 00:37:54,207 to the veeam's, uh, passwords. 610 00:37:54,207 --> 00:37:58,147 And those admin passwords has a separate independent, so it, it has 611 00:37:58,147 --> 00:38:04,207 that, uh, ability to, uh, operate, uh, without a risk of uh, uh, you 612 00:38:04,207 --> 00:38:06,067 know, global password type settings. 613 00:38:06,577 --> 00:38:07,897 And, um, and then of course, the 614 00:38:07,897 --> 00:38:08,197 last 615 00:38:08,232 --> 00:38:09,162 Prasanna Malaiyandi: I like that part, 616 00:38:09,787 --> 00:38:10,177 David Cerf: Yeah. 617 00:38:10,207 --> 00:38:10,597 Yeah. 618 00:38:10,812 --> 00:38:11,592 Prasanna Malaiyandi: Yeah, I know, I know. 619 00:38:11,592 --> 00:38:15,132 We always talk about Curtis about, yeah, don't put your backup servers on 620 00:38:15,137 --> 00:38:18,282 the same ad right as everything else. 621 00:38:18,282 --> 00:38:18,972 So I'm glad that 622 00:38:19,000 --> 00:38:19,900 W. Curtis Preston: Separation of powers, 623 00:38:20,017 --> 00:38:20,407 David Cerf: No, no. 624 00:38:20,407 --> 00:38:21,307 Post-it notes. 625 00:38:21,552 --> 00:38:22,242 Prasanna Malaiyandi: and separating it. 626 00:38:22,602 --> 00:38:22,902 Yeah. 627 00:38:23,022 --> 00:38:25,542 Well I'm glad you guys are going a step further and not even having 628 00:38:25,542 --> 00:38:28,602 like the normal being passwords as this authentication mechanism, 629 00:38:28,602 --> 00:38:31,992 cuz you really do want that more secure than everything else. 630 00:38:32,707 --> 00:38:33,277 David Cerf: Absolutely. 631 00:38:33,397 --> 00:38:33,727 All right. 632 00:38:33,787 --> 00:38:35,347 This is, this is your last resort. 633 00:38:35,347 --> 00:38:35,527 Right? 634 00:38:35,527 --> 00:38:38,557 And we're, and that's really the key is that we, why do we back up? 635 00:38:38,557 --> 00:38:43,207 We pack up because only when we absolutely need that data and if they take that down. 636 00:38:43,477 --> 00:38:47,797 So the blocky provides that additional layer of security and protection. 637 00:38:48,367 --> 00:38:52,867 Um, and it works, of course, uh, you know, we, we have the ability to single 638 00:38:52,867 --> 00:38:57,487 site, multi-site and, and, um, so it provides a, this really simple way 639 00:38:57,487 --> 00:39:01,597 for whoever is managing either the, the IT stack or the, you know, the 640 00:39:01,597 --> 00:39:04,297 security stack to add a layer into. 641 00:39:04,952 --> 00:39:06,322 A product that is fantastic, right? 642 00:39:06,322 --> 00:39:10,562 I mean, Veeam Veeam is uh, you know, proven globally and customers 643 00:39:10,567 --> 00:39:13,282 love it, but now they can have that additional protection. 644 00:39:14,621 --> 00:39:16,581 Herbert Grau: Maybe one more comment from my side. 645 00:39:16,581 --> 00:39:21,081 People ask me, what's the performance impact if I have blocking installed? 646 00:39:22,675 --> 00:39:23,335 W. Curtis Preston: That's an important 647 00:39:23,811 --> 00:39:23,991 Herbert Grau: Yep. 648 00:39:24,381 --> 00:39:28,911 The answer is, while writing and readings, we don't do nothing. 649 00:39:29,721 --> 00:39:33,771 It's not like a virus can always, always holds the process and then 650 00:39:33,771 --> 00:39:35,541 does not recognize the bad guy. 651 00:39:36,381 --> 00:39:39,141 So we do nothing while writing and reading. 652 00:39:39,891 --> 00:39:45,981 When it's deleting or modifying, we hold the process and check it 653 00:39:46,491 --> 00:39:48,241 because that's the purpose of blocky. 654 00:39:48,741 --> 00:39:55,671 And then we have, um, uh, maybe a two to 3% overhead while deleting and modifying. 655 00:39:56,331 --> 00:39:57,591 And that's a cool combination. 656 00:39:58,122 --> 00:40:00,192 Prasanna Malaiyandi: Well, and especially because reading, or, sorry, 657 00:40:00,612 --> 00:40:06,312 deleting and modifying isn't your predominant uh, uh, operation right. 658 00:40:06,411 --> 00:40:07,041 Herbert Grau: Of course not. 659 00:40:07,371 --> 00:40:09,741 And if this happens, you want somebody to check 660 00:40:09,912 --> 00:40:10,302 Prasanna Malaiyandi: Yeah. 661 00:40:11,375 --> 00:40:11,655 W. Curtis Preston: Yes. 662 00:40:11,845 --> 00:40:12,455 Agreed. 663 00:40:13,375 --> 00:40:18,955 Now, one, one question that I have, uh, this will be my, my toughest question. 664 00:40:19,345 --> 00:40:21,985 Is there a way to defeat this product? 665 00:40:22,615 --> 00:40:27,685 So if I have admin on the box, What am I able to do? 666 00:40:27,685 --> 00:40:29,965 I know you, if the product is installed, 667 00:40:30,051 --> 00:40:31,191 Herbert Grau: you, if you want an honest 668 00:40:31,191 --> 00:40:31,641 answer, 669 00:40:31,641 --> 00:40:33,831 I can give you the honest answer. 670 00:40:33,891 --> 00:40:38,571 An admin can destroy the whole Windows machine 671 00:40:40,015 --> 00:40:40,435 W. Curtis Preston: Right? 672 00:40:41,511 --> 00:40:44,631 Herbert Grau: and that's not possible to avoid. 673 00:40:44,841 --> 00:40:48,861 Neither from Veeam, not from Crau, not from Microsoft today. 674 00:40:50,245 --> 00:40:50,635 W. Curtis Preston: Right. 675 00:40:50,905 --> 00:40:52,285 That's a pretty honest answer. 676 00:40:52,615 --> 00:40:57,115 Um, does that mean also that they could uninstall the product 677 00:40:57,591 --> 00:40:58,671 Herbert Grau: No, that's protected. 678 00:40:58,851 --> 00:41:00,231 The uninstall is protected. 679 00:41:01,495 --> 00:41:01,505 W. Curtis Preston: Okay. 680 00:41:02,485 --> 00:41:04,375 How, how, 681 00:41:04,551 --> 00:41:06,741 Herbert Grau: years ago, and that's protected 682 00:41:08,275 --> 00:41:11,215 W. Curtis Preston: How, um, I don't want to get into secret 683 00:41:11,215 --> 00:41:13,195 sauce, but how, in what way? 684 00:41:13,195 --> 00:41:14,755 Like how, how do you protect that 685 00:41:14,901 --> 00:41:16,641 David Cerf: You need, you need a password to go back. 686 00:41:16,641 --> 00:41:20,211 And so I mean the, I think the real security here is if you have, if 687 00:41:20,211 --> 00:41:23,601 you have the admin and they blow the box away, they blow the box away. 688 00:41:23,631 --> 00:41:26,271 I mean, so we're de that's a physical security issue potentially. 689 00:41:26,271 --> 00:41:26,421 Right. 690 00:41:27,301 --> 00:41:33,061 W. Curtis Preston: I think what I'm, what I'm concerned about is not somebody who's, 691 00:41:33,211 --> 00:41:37,141 you know, so we've got a malware in there, we've got a, a bad actor in there, and 692 00:41:37,371 --> 00:41:42,451 they're trying to surreptitiously access data that they're not supposed to access. 693 00:41:42,781 --> 00:41:43,141 Right. 694 00:41:43,146 --> 00:41:49,231 So they would want to disable, um, this, this tool, and it sounds like that 695 00:41:49,236 --> 00:41:52,621 without the username and password from that tool, they wouldn't be able to do 696 00:41:52,773 --> 00:41:53,333 David Cerf: Right. 697 00:41:53,403 --> 00:41:54,373 I, I mean, so. 698 00:41:54,611 --> 00:41:57,116 W. Curtis Preston: because blowing up the box, they would, they 699 00:41:57,116 --> 00:41:59,546 would obviously show their hand. 700 00:41:59,876 --> 00:42:00,206 Right. 701 00:42:00,211 --> 00:42:01,766 So they're not likely to do that. 702 00:42:02,066 --> 00:42:04,646 What they're likely to do is to try to disable anything that's 703 00:42:04,646 --> 00:42:06,146 trying to block their access. 704 00:42:06,473 --> 00:42:10,463 Herbert Grau: Maybe, maybe one interesting point is that we have sold blocky also 705 00:42:10,468 --> 00:42:12,683 to one very large customer in Stuttgart. 706 00:42:13,613 --> 00:42:18,293 Which has 100, uh, IBM backup server from tsm. 707 00:42:18,298 --> 00:42:19,583 Now spectrum scale. 708 00:42:21,143 --> 00:42:22,853 And that's a huge environment. 709 00:42:24,203 --> 00:42:27,623 And this is a corporate license we sold here. 710 00:42:27,633 --> 00:42:29,393 We're very, very proud about this. 711 00:42:30,653 --> 00:42:35,333 Uh, you may understand that we cannot name , give names out because in, 712 00:42:35,333 --> 00:42:39,443 in this ransomware world, nobody wants to read his name anywhere. 713 00:42:40,193 --> 00:42:45,383 Uh, but the point is that in the deep in in, in the TSM world, I still call it tsm. 714 00:42:46,253 --> 00:42:48,743 Um, and Curtis, you know, maybe you two 715 00:42:49,167 --> 00:42:49,797 W. Curtis Preston: Yeah, me too. 716 00:42:50,153 --> 00:42:52,343 Herbert Grau: the old guy, as we call, we still call it tsm. 717 00:42:52,587 --> 00:42:55,987 W. Curtis Preston: I still call, I still call a dsm, by the way. 718 00:42:56,333 --> 00:42:58,193 Herbert Grau: who who knows that, you know. 719 00:42:58,913 --> 00:43:03,923 But, uh, in the TSM world, there's also always a DB two coming with a product. 720 00:43:05,157 --> 00:43:05,767 W. Curtis Preston: Mm-hmm. 721 00:43:05,818 --> 00:43:11,543 Herbert Grau: cool from my side is that we can also protect the DB two data. 722 00:43:13,667 --> 00:43:13,787 W. Curtis Preston: Hmm, 723 00:43:14,318 --> 00:43:19,658 Herbert Grau: Which opens potentially a market to applications that 724 00:43:19,658 --> 00:43:22,208 will also protect the database 725 00:43:22,379 --> 00:43:22,869 Prasanna Malaiyandi: Yeah. 726 00:43:23,228 --> 00:43:23,678 Herbert Grau: data. 727 00:43:24,007 --> 00:43:24,677 W. Curtis Preston: exactly. 728 00:43:24,848 --> 00:43:26,348 Herbert Grau: That's our next step. 729 00:43:26,408 --> 00:43:27,218 Potentially 730 00:43:27,729 --> 00:43:31,569 David Cerf: and by the way, that customer also had Veeam, oh, I'm sorry. 731 00:43:32,218 --> 00:43:32,378 Herbert Grau: Hmm. 732 00:43:34,398 --> 00:43:34,958 Sorry. 733 00:43:35,139 --> 00:43:37,749 David Cerf: I, I was just gonna say that, that same customer, not just that 734 00:43:37,749 --> 00:43:41,199 they have tsm, but they also have Veeam. 735 00:43:41,589 --> 00:43:42,759 So they're, they're happy, 736 00:43:43,418 --> 00:43:45,788 Herbert Grau: customers which have, which you have tsm, have 737 00:43:46,238 --> 00:43:46,478 Other 738 00:43:46,499 --> 00:43:47,189 Prasanna Malaiyandi: Other things. 739 00:43:47,189 --> 00:43:47,519 Yep. 740 00:43:48,399 --> 00:43:50,889 David Cerf: So they're, so they're now, now that they're secure on their 741 00:43:51,009 --> 00:43:55,329 tsm, it, add, add, add, the additional protections to their Veeam is where 742 00:43:55,329 --> 00:43:57,159 they're heading next, um, as well. 743 00:43:57,214 --> 00:44:00,724 W. Curtis Preston: Well, I, I wanna, I want to thank you for, uh, you know, this 744 00:44:00,729 --> 00:44:02,314 has been a good, really good discussion. 745 00:44:02,314 --> 00:44:05,254 I, I've learned more about the, you know, obviously about all 746 00:44:05,254 --> 00:44:06,364 of the products that you do. 747 00:44:06,694 --> 00:44:09,874 We've focused in on the end here on, on Blockie for Veeam. 748 00:44:10,294 --> 00:44:14,524 Uh, and I, I think you've got a tremendous potential market. 749 00:44:14,574 --> 00:44:17,454 Veeam has a lot of customers, and every one of 'em has a window 750 00:44:17,459 --> 00:44:18,624 server that needs protecting. 751 00:44:18,624 --> 00:44:25,524 So, uh, I, I wish you, uh, the best of luck and, um, thanks so much for, 752 00:44:25,854 --> 00:44:29,694 for, for standing, for allowing us to stand between you and a beer, Herbert 753 00:44:29,754 --> 00:44:33,164 Herbert Grau: Yeah, actually it's a bottle of wine today. 754 00:44:34,286 --> 00:44:34,556 Prasanna Malaiyandi: Even 755 00:44:34,556 --> 00:44:34,946 better. 756 00:44:35,454 --> 00:44:35,994 W. Curtis Preston: All right. 757 00:44:36,594 --> 00:44:39,024 Something from the Rhine region perhaps. 758 00:44:39,139 --> 00:44:41,539 Herbert Grau: Uh, could be Ryan, could be Mosel. 759 00:44:41,629 --> 00:44:43,459 You know, we have some valleys 760 00:44:43,459 --> 00:44:43,729 here. 761 00:44:45,409 --> 00:44:47,449 W. Curtis Preston: Well, well, thank, thanks a lot everybody 762 00:44:47,454 --> 00:44:49,399 for, for being on the podcast 763 00:44:50,790 --> 00:44:53,310 David Cerf: Thank you for having us for, appreciate the discussion. 764 00:44:53,744 --> 00:44:55,124 Herbert Grau: Thanks, Curtis. 765 00:44:55,334 --> 00:44:55,694 Thanks. 766 00:44:57,066 --> 00:44:57,546 Prasanna Malaiyandi: you all. 767 00:44:57,546 --> 00:44:57,666 Yeah, 768 00:44:57,876 --> 00:44:58,386 great. 769 00:45:00,019 --> 00:45:00,799 W. Curtis Preston: absolutely. 770 00:45:00,859 --> 00:45:06,559 And again, as always, we'll remember to, uh, thank our listeners and uh, be sure to 771 00:45:06,559 --> 00:45:08,929 subscribe so that you can restore it all.