SNIA Experts on Data
Storage Technology Update: SNIA and SCSI Trade Association Join Forces
Uncover the alliance of SNIA and SCSI Trade Association in a conversation with Cameron Brett from the SNIA STA Forum and Dr. J Metz from the SNIA Board of Directors. They'll guide you through the compelling history of these two organizations, their recent unification, and the benefits for members and end-users. Learn how the STA Forum, while being a part of SNIA, continues to maintain autonomy, and what this new union means for the future of storage technology.
SNIA is an industry organization that develops global standards and delivers vendor-neutral education on technologies related to data. In these interviews, SNIA experts on data cover a wide range of topics on both established and emerging technologies.
About SNIA:
Uncover the alliance of SNIA and SCSI Trade Association in a conversation with Cameron Brett from the SNIA STA Forum and Dr. J Metz from the SNIA Board of Directors. They'll guide you through the compelling history of these two organizations, their recent unification, and the benefits for members and end-users. Learn how the STA Forum, while being a part of SNIA, continues to maintain autonomy, and what this new union means for the future of storage technology.
SNIA is an industry organization that develops global standards and delivers vendor-neutral education on technologies related to data. In these interviews, SNIA experts on data cover a wide range of topics on both established and emerging technologies.
About SNIA:
Welcome to the SNEA on Data Podcast. Each episode highlights key technologies related to handling and optimizing data. This is SNEA on Data Podcast, snea and STAY in storage consolidation, where you'll hear from Cameron Brett, snea's STAY forum chair, and Dr Jay Met, chair of the SNEA board of directors.
Speaker 2:Discuss the SCSI Trade Association joining SNEA and what this positive development means for the storage industry Today we're going to talk about the fact that SNEA and the SCSI Trade Association have come together, join forces, how we describe it, but perhaps, before we get on to that, the implications of that, how that's all working. It'd be good to understand a little bit about the SCSI Trade Association the history, what it's achieved, what it does, etc. Up until this point.
Speaker 3:Yeah, thanks, phillip. The SCSI Trade Association has been around since the mid-90s and it's been the marketing arm of the T10 committee that defines the specifications for the 11 or so generations of SCSI and SAS technology that we know today. It combines a number of member companies within the industry, from SSDs to systems makers to cables, so it's a large part of the ecosystem. So, yeah, we've been around promoting the latest generation of technology, 24ghz, to date.
Speaker 2:Okay, and then I guess, as I said, we're talking about the fact that the two of you, the SNEA and the SCSI Trade Association, have come together. It'd be interesting to understand unless there are any secrets you can't reveal how did that come about? What's the thinking behind it? So what were the steps to joining forces? I'm amused at the thought of having secrets.
Speaker 4:That's kind of funny. Go ahead, Cam.
Speaker 3:Well, just one little known fact about SCSI technology. I mean it's very, very pervasive in the world and a lot of people don't even know that they use it. It's kind of been this kind of cryptic technology that's been for high-end enterprise systems, even in USB devices like your thumb drive uses SCSI commands. Other technologies like Fiber Channel 1394 also utilize SCSI commands. But the joining of forces between the SCSI Trade Association with SNEA began maybe about a year and a half ago. We had some kind of casual discussions which grew into more focused discussions on does it make sense for the state membership as well as SNEA, to have the state become part of SNEA? In April earlier this year, after many months of discussions, the SNEA leadership voted to approve the joining of SCSI Trade Association and in May of this year we officially became a part of SNEA as the SCSI Trade Association Forum.
Speaker 2:Okay, and in terms of the forum, because obviously SNEA already has four, if that's the correct plural of forums. But what is the status of the STA forum? Because clearly the other forums, if I'm right, are all being formed from within SNEA. This one is being brought from outside and clearly you had your own autonomy. So do you have any slightly, or will this forum have any slightly different status, bit more autonomy, et cetera, or just how it will appear within the SNEA organization?
Speaker 4:As it turns out, snea had a rather broad charter for being able to incorporate new organizations and existing organizations into the fold. So what we've been able to do is set up a structure inside of SNEA for technical work, development, marketing work, development, educational work, development, and then we've kind of had this special category for affiliates where you have these kind of all-in-one entities like. Sff, for example, is an example of an affiliate. The DNA data storage alliance is another affiliate. However, the STA group fits very, very well inside of the forums and initiatives category. It's charter, it's the work that it does, the relationship that it has with T10.
Speaker 4:All of these things are perfectly aligned with the forum structure that we've currently got, so there was really no need to create something new for it.
Speaker 4:They still I see they, but it's a little bit difficult for me to try to figure out the right pronouns here but the STA forum effectively still has its own charter and is still going to do the work that it's currently doing. The only real main difference that it's going to fall under the broader rubric of the SNEA family. So one of the things that we found to be encouraging by this was that, as we start to figure out the way that the relationships between the different organizations inside of SNEA and outside of SNEA become more prominent and more important, the relationships themselves, that is. It makes a great deal of sense to have something like the stay organization come in and help be a bedrock for that. So I think it's a win-win situation for both organizations because we've been trying to get inside of SNEA for a while now a stronger internal bond between the different groups, as well as an internal, an external bond with other organizations and state. That's right perfectly in that sweet spot, so we're very happy about it.
Speaker 3:From a structural standpoint we are now a forum and, as SNEA organizational structure has it, I will act as a chairperson for the organization. But we still have autonomy within the group itself to kind of operate the way we were before. So we have a board of directors and officers within that board of directors, a marketing work group as well as a general membership. So we kind of get the best of both worlds. We can fit nicely in, as Jay mentioned, within the organization from a content standpoint and we can still operate similarly to the way we have been for many years.
Speaker 2:Okay, and Jay, in answer to that, you alluded, I think, to the potential benefit, at least to members within SNEA, but it'd be good to understand what you see as the joint, whether it's one of your, both of you, the joining together. What are the benefits, both for the member companies of both organizations and potentially I guess there was some crossover, membership crossover but also for the end users? I mean, what will they notice in terms of the benefits that the two of you combined will bring?
Speaker 4:I think what the members are going to find is that, with the focus on the ability to have stronger relationships between the groups, they'll have resources and expertise that they wouldn't normally have available to them at their fingertips. Right, an official relationship between a state organization and the other groups, whether they be technical or marketing or educational focus. It helps simplify a lot of things that we're trying to do. So it's important to note that over the last three years, which is the time period that I've been chair of the organization, we've been trying to get SNEA to be more holistic. You know, in the in the hip-ass history of the way SNEA has worked, which worked very well for the environment in which that it, you know, operated, we've tried to shift to more of a synthesis of the different kind of groups. Right, we want the technical groups and the educational groups and the marketing groups and, you know, the affiliates, to all act in concert with each other. You know, just like any kind of orchestra, where you've got a you know a woodwind section, you've got a brass section, you've got a percussion section, they can't all be playing different songs at the different, you know, at the same time. So you know where we start to work on.
Speaker 4:We have this really robust knowledge base that we've been able to share internally as well as externally, and I think that what the members are going to find is that they're going to be able to have even more experience at their fingertips. One of the things that we're looking forward to, as a matter of fact, is we're going to be completely revising and revamping the way that we present ourselves externally to the world. You know we're going through that very process and you know our websites and our events and our conferences. They're all going to be focused on this kind of of synthesis, of integration for all of the different. You know events and initiatives and concepts that we all, you know, know and love in terms of storage, and you know data relationships.
Speaker 2:And in terms of the sorry, sorry Kamu, do you want to do something?
Speaker 3:Yeah, if I might add. Yeah, I mean definitely. The members of the Schoesie Trade Association Forum will now have additional access to a range of SNEA technologies and initiatives, as Jay noted. But we're also hoping that we might be able to increase the growth of the state membership. Of course, with SNEA being a very large organization that does a lot of events, the state members also can have added participation in SNEA events.
Speaker 2:And in terms of just the practicalities, I suppose, such as they are in terms of personnel, I mean websites, jay, I think you said there anyway you're sort of re-evaluating what SNEA is doing in that way anyway. But just the transition, if we want to call it that in the same way, two companies merge. I hope it won't be as painful as some of those. But in terms of the practicalities, what is going to be done over what sort of timeframe?
Speaker 4:Well, the way I see it is that this is sort of an iterative process of evolution. So where we get into trouble is when people fail to understand that there are shoulders of giants that they can stand on and they try to reinvent the wheel. I'm going to try to see if I can mix more metaphors into this, so bear with me. But where things have gotten to be messy inside of SNEA and what we're trying to fix is that we have a lot of hodgepodge. Organic growth over the last decade and a half or so I mean we've been around for 25 years but as we started with these different initiatives and so they've all been kind of separate and isolated and not really well connected. But this kind of is anathema to our current mission. We want people to be able to easily find things and, quite frankly, the SNEA collective information dump is an awful lot like going into a vintage store. You know there's good stuff in there somewhere, but you've really got to have to look for it and you have to be willing to be in the right mindset to do that. We want to try to fix that. We want to make sure that the relationships between the different organizations and the initiatives and the technical and the education library and all that kind of stuff is very, very easy to find and the relationships between them is easy to understand. So that's the work that we're trying to do, and I think one of the things that's particularly I'm particularly enthusiastic about is that the wealth of knowledge that the organizations of stay and the organizations of SNEA, which have these decades of background and history, are going to be something that people can use without having to reinvent that wheel that I was talking about earlier.
Speaker 4:I mean, most of the time is people like well, I'll just go ahead and do whatever I have to do because nobody's ever done this before and it turns out it was done 30 years ago. So I'm trying to personally, I'm trying to encourage people to do kind of a yes and approach yes, you're doing this and it relates to something else that someone else is doing Right. So, yes, we're doing DNA storage and it relates to some of the new media types. Yes, we're doing SCSI and it relates to, as Cameron was talking about, usb connectivity and external hard drives or whatever winds up being right. That yes and approach must have the ability to let you follow it If we say yes and then you can't do the and part, then it doesn't help. So that's where we're trying to go with that. That's where we're trying to go with that.
Speaker 2:Okay, and in terms of sorry have I interrupted. I apologize.
Speaker 3:Cameron, Will you get to add to that, or yeah, I was just going to add, at least from a technology standpoint, the Schleswig Trade Association Forum is going to continue to work closely with the T10 as the marketing arm of the technology. From the membership standpoint, we're actually hoping to grow that the number of members now that were accessible to the larger SNEA membership and the overall structure and the way we operate it is going to remain the same as it did before since before we joined forces. But I definitely see a lot of the benefits of being part of SNEA as a larger organization is definitely going to come into play and benefit the staff forum members greatly.
Speaker 2:Okay, and I know Veer, jay and one or two others. I try and keep focus up to date with what the SNEA is focused on, but it'd be good to understand similarly from Stays' point of view, what sort of projects, what work are you currently focusing on within the organization?
Speaker 3:Okay, yeah, again from a technology standpoint, 24gig SAS has been introduced a few years ago, but the adoption within the industry is now starting to become a little bit more mainstream. And in addition to that in the pipeline we have what we're calling 24GIG plus SAS, which is not really as much of a speed turn like historically the generations have done, but it's rather looking at some of the key features that improve management, reliability and security. So SAS, of course, is a very mature technology and there's not going to be any large, big earth-shattering innovations, but more it's kind of refinement to help keep things very stable and secure and reliable.
Speaker 2:Okay, and although I said, jay, you do keep me up to date with what you do, and I know some of your colleagues do, is there anything? What a highlight is the what SNEA is doing at the moment or in the pipeline, or do we know about everything at the moment?
Speaker 4:Oh well, I mean, I personally think that the state announcement is a relatively large one. So I mean, we're really kind of excited about that, and we haven't even I don't even think we've had a chance to really touch on the areas of collaboration that we can really truly do. We are excited with some of the things that have been happening. I think one of the things that I'm particularly keen on in the upcoming months is that we've been, over the last year and a half or so, been really broadening our touch points. So at the Storage Developer Conference last year, we not only had SCSI presentations, but we had CXL presentations, we had blockchain presentations, we had management presentations, we had DNA storage, computational storage, memory movement. We cover so many things, and the reason for that is because there is and I've been saying this for a long time there is a collision course between storage and memory, and right now we're already at a point where many of the things that people are talking about in terms of storage and memory are already kind of munched. We've already got the work inside of NVM Express, which is a fantastic organization, to look at extension of memory into SSDs. We have the ability to do data movement with a smart data accelerator interface for SDXI inside of ASNIA, which is a software-based portable data mover technology. Well, that opens up memory as a semantic for storage systems. And then, when you add on to these kind of granular approaches to accessing data, it doesn't really matter in a conceptual sense whether it happens to be the load store addressable, byte addressable, deword addressable, block addressable, file addressable. These are all kinds of variations on that particular theme, even though the semantics do wind up changing over time. But they're getting very close together because our margin for error and latency is going down.
Speaker 4:It used to be that memory was a 200-nitosecond or 100-nitosecond latency and storage was 8 milliseconds. We're talking orders of magnitude difference and now not anymore. Now we're still talking about CXL memory locations that are just outside of the latency budget and they're being put inside of across the CXL link, psb, cie links and so on and so forth, and they're all being put into some sort of persistence. Well, once you start talking persistence, you start talking about systems of reliability, systems of management, systems of protection and security systems that don't normally have a history inside of memory but do inside of storage.
Speaker 4:As we start to go through these kinds of initiatives, we're just scraping the surface of what can be done when you start to rethink your questions, redefining the questions, and this goes for memory, it goes for the processing, it goes for the storage world, goes for the networking, all the things we've been talking about for computational storage and so on, about moving the compute processes closer to the data, whether it be on a network or whether it be local. All of these things have consequences, and we're very excited about the fact that this is kind of like. I've been kind of likened to the transformers toys, where you have a particular toy or a particular transformer that does its own thing, but then, when it's formulated in the right way, it combines with other transformers to create a much larger and more powerful one. I think that that's kind of where we're looking to go here, too, is that every one of these different technologies is valuable on its own, but when you combine it with something else, it's much more powerful than just the sum of its parts.
Speaker 2:Okay, and in terms of I mean clearly we're talking about the stay today and what the potential of that is but would it make sense? Or are you open to? I know you collaborate work alongside other different organisations, but could you see a world where, instead of having three or four, five different storage focus bodies, it might make sense to all come together under the one umbrella? I mean, I was on the you know, FCIA, for example. I know them historically or just on the website today, before we came here, and they don't seem to have a lot of member companies, but I know they do. Standards work, but do you think there might be a logic in all coming together? Or you just see if it happens organically, great, but you're not sort of looking to bring everyone together?
Speaker 4:I think it's an interesting question. I have a long history with FCIA. I was on the board of directors for a long time and I would also participate in a member of T11. So you know this kind of a parallel between stay and T10 and the FCIA and T11. So there is a logic to that. I think part of it has to do.
Speaker 4:The thing to remember is that there's more to it than just saying we're going to join forces, right, there are licensing requirements, ip requirements. You know the standards organizations have to align. I mean, for example, you know, t11 is a part of insights, for instance, right, and the IP structure that's seen is typically ISO. So those are those kinds of needles have to be threaded properly. So it's not a trivial nature to go forward. On the other hand, to your point, there is a logic to you know, including these kinds of technologies where they can take advantage of that broad synthesis that we were talking about before, add their instruments to the orchestra, so to speak. So it's not there's no, yes or no answer to this, it's just okay. This is one of those things that we have to try to figure out if it's a good thing to do or not. But we haven't had those conversations yet, to be honest, but it doesn't mean that it can't happen.
Speaker 2:Okay, and first of all I've got you both here. It'd be remiss of me not to ask her to finish with, at least in terms of the storage industry, if you can. You know, happy both to answer maybe the number one challenge and I know, joe, you've talked a little bit beforehand, but the number one challenge and whether you define the challenge also as an opportunity. But what do you think is the storage industry? Yeah, what's coming down the tracks that maybe worries you a bit and or excites you, if that's reasonable to ask.
Speaker 4:All the Cameron talk versus I've been blathering about for a while.
Speaker 3:Thanks. I mean there are lots of challenges, you know. One of the bigger ones is being able to accommodate and manage and protect all the growth in data that's being created these days. A lot of it is temporarily, temporarily stored and a lot of it is permanently stored. I work at a flash memory maker and an SSD maker, kyokusha and it's a lot of times it's difficult to keep up with all the demand.
Speaker 3:But also security from a data protection standpoint as well, as you know kind of malware and and you know bad elements protecting data from that perspective. So I see, definitely like encrypted. Encrypted, encrypting data and other data protection mechanisms are definitely some of the bigger challenges, at least that we see from an opportunity standpoint, you know, of course there's plenty of room for innovation and new applications and one of the things that we're starting to see, you know, at least from a technology standpoint SCSI SSDs are going to be going up into the International Space Station. So this, this very much for proven technology, is going to be expanding into space. So we're very excited to see that happen, jay.
Speaker 4:I want to take that, your answer, cam, and kind of turn it a little bit sideways, because I think there's a, there's a fundamental question that needs to be addressed first, or at least, at least in concert, and I think that the biggest challenge that the industry has is a fundamental misunderstanding of what storage actually is to be, just to be, to be blunt, as I've gotten to work so Cam works for Keoghsha and I work for a processing company right, I've worked for AMD. So one of the things that I've noticed in the, in the, you know, in my travels, is that when people who are not storage people think of storage, they think of capacity, they think of the actual media, whether it be hard drives or SSDs or DNA data storage right, they think of the actual medium on which the storage is is placed, or they think of how much you can put there. That is remarkably short sighted. Storage is a system. Storage is an approach to saving, preserving and making data available, because ultimately, you can't just put something somewhere and expect it to still be there when you come looking for it again. So all the systems that have to be put in place, all of the you know, the security and the structure and the reliability, all to give you back that bit. That's storage, not just where the bit resides.
Speaker 4:And this is probably the biggest challenge that I think the industry has, which is, when we talk about these things, about the things that we're doing, we're we're basically casting our voices into the ether if we're using a different definition than the rest of the world thinks.
Speaker 4:I can tell you from personal experience that the networking people who think of storage is some sort of device at the other end of the wire, the processing people you know the ones that I work with in my own company think of storage as not my problem. We do all the hard work and it's up to them to have the data stored and I think ultimately, at the end of the day, we wind up with this you know this issue that we need to properly convey the significance and the consequences of what data protection, data retention, data management, data movement actually means, because all of that is what storage is and it's not a hard line that goes around the concept. It's a fuzzy line and the Venn diagrams overlap quite a bit, and that's ultimately very true when the memory collides with storage, when the networking collides with storage, when the computation collides with storage, because thinking of it as solely as a place to store data is going to cause more headaches and more trouble than people even realize.
Speaker 2:And I'm just fun to follow that up because I know I've taken a lot of your time, so I won't be too much of a problem. But do you think that is potentially because, again to the outside world, storage, when it first started, came on the scene, if you like, as a topic? It was complicated, it was exciting, a lot of people doing this thing and over years it's decades, whatever it's been seemed to be somewhat commoditized, rightly or wrongly, and is that why people therefore don't take it as seriously and understand the nuances of what's going on under the covers? If you like.
Speaker 4:Well, I guess I'll depend on how cynical you want to be to the answer to that question. I mean, if you look at the history of computation all the way back from the 40s, when we were start talking about the massive mainframes and then the many computers and so on and so forth, the one thing that was not democratized in a way that others have was been storage right. So the compute became democratized when they moved from mainframes to x86 and the pizza box servers. Networking became democratized when ethernet became so flexible and pluggable where you could do ethernet-based storage, ethernet-based networking solutions in a small scale and large scale. And then of course, you had the ARPANET and then ultimately the internet.
Speaker 4:The ability for a cottage industry to come up and support both the servers and the network, but not storage, was probably what made storage very successful in the 90s and the early 2000s, but ultimately wasn't a long-term strategy because it became this kind of black art. And then what wound up happening was somebody said oh, you know what, I could put storage inside of my servers and treat it like it's a server. And so then we have this whole decentralized, disaggregated storage. We have software-defined storage that came out of the software-defined networking did not come out of regular storage history, and so we kind of wound up with this. Well, storage was 25 years ago and it's not, so we were kind of a victims of our own approach to doing this, I think, and we don't really have anybody else to blame but our own inability to properly communicate to the audience that we needed to. Well, okay, so it's not too late, right, it's never too late because we still have a lot of experience in this industry that can be distributed so that the server people and the networking people don't have to reinvent the wheel.
Speaker 4:But that's gonna require outreach on both sides and a lot of the storage people have to recognize that there are limitations to our own understanding, right. And that kind of arrogance isn't just indigenous to storage, right, I mean, it exists everywhere, in all different disciplines. So my personal goal in this is to try to make the outreach as easy as possible for anybody who's curious about this to be able to find the information they need. Because in my mind, the most approachable technology is going to win. The easiest to put into place, the easiest to provision, the easiest to deploy, that's going to be the one that's going to be the most important and I think one of the things that we have an opportunity to do is do that kind of outreach, do that kind of integration with other organizations that may not historically be storage oriented or, conversely, may not typically be memory oriented, right? So those aspects that were all about redefining the questions, not coming up with better answers, because sometimes the answers already exist, if you just look at it a different way.
Speaker 2:Okay. So it sounds to me I don't want to put words in my mouth but it sounds to me that the coming together of your two organizations is like the early stage of a bigger push to put storage back on the map where it deserves to be. So, but anyway, for now at least, because I've taken a lot of your time, both Jay and Cameron, I really appreciate you giving me the time and sharing some fascinating answers. So, gentlemen, thanks very much for your time, thank you.
Speaker 4:That's great, because if you don't like those answers, we've got more.
Speaker 2:I'm sure we will speak again. Thanks again, folks. All right, take care, y'all.
Speaker 1:Thank you for listening. For additional information on the material presented in this podcast, be sure and check out our educational library at sniaorg slash library.