Database Virtualization: A Stepping Stone to Virtualization 2.0

By Derrick Harris

August 21, 2008

In this interview with On-Demand Enteprise, Xkoto Chief Strategy Officer and Co-Founder Albert Lee discusses the ins and outs of the database virtualization market, as well as the future direction of Xkoto’s one-of-a-kind Gridscale solution. Currently supporting only IBM’s DB2 database, Gridscale will support SQL Server within the next 30 days. Lee was a panelist at the Next Generation Data Center conference, discussing with other virtualization vendors the advent of Virtualization 2.0.



ON-DEMAND ENTERPRISE:
Can you explain the concept of database virtualization, in terms of the technology and issues it addresses for businesses?

ALBERT LEE: The way the technology works is we provide a software appliance approach. We sit transparently between the applications and a pool of databases that we’re managing, which make that pool appear like one database. That’s the virtualization magic that goes on. The application connects to our Gridscale server thinking that it’s connecting to its database and, instead, we’re able to load-balance queries down to that pool and also keep all the copies of data in that pool asynchronously updated so they can survive any kind of failure and support true online maintenance without taking applications down. You’ve got a complete independent and consistent copy of data, [and] typically, customers will put those copies in that pool in different datacenters so they have got true location independence, disaster avoidance, continuous availability and, as I mentioned, with that pool approach we can now scale that workload across and speed up performance. We’ve shown 85 percent or higher scalability when we’re doing a lot of querying. That is, in a nutshell, the architecture and how this all works.

ON-DEMAND ENTERPRISE: You used to describe the technology as “data virtualization.” Why the switch?

LEE: We see this as being a clarification to the marketplace. … Data virtualization is a perfectly good label, but it also means many things to many people and many vendors. We wanted to make it absolutely clear to our customers and our partners that what we’re virtualizing in that data tier is the database itself, rather than a data stream, or some kind of federated data solution, information fabric approach, or some hybrid between a cached or federated shared-memory database or spindle-based database. Those are all real, cool notions that exist in data virtualization, and we play with those different models, but we wanted to make it really simple for people to understand where in that whole data virtualization universe we live, and it’s the database.

ON-DEMAND ENTERPRISE: How does database virtualization fit into the overall virtualization picture? For example, how does it fit into a server virtualization strategy?

LEE: The idea of getting server consolidation and benefits, and driving down box counts and increasing the number of environments under management by each admin, those were the driving forces to adopting hypervisors in the so-called Virtualization 1.0 use case. That’s come along real well, and according to that we heard on the [Virtualization 2.0] panel, it’s really at the 10 percent adoption rate and increasing every day.

How we play in those environments is as IT people put hypervisors onto their physical boxes, they have this new challenge of being able to put databases and enterprise applications that access those database onto hypervisors, which divvy up the box but, as result, can leave intensive applications like databases starved for power. Having our Gridscale solution bringing about this scaling effect by aggregating the capacity across multiple VMs or physical servers, or a combination thereof, really speaks to the need that has been created by the hypervisor. That has been a pretty strong value proposition for adopting database virtualization once you’ve gone the first step of adopting server virtualization.

ON-DEMAND ENTERPRISE: Speaking of Virtualization 2.0, and considering the relatively small numbers of virtualized servers, is database virtualization part of that “version,” or is it really a Virtualization 3.0 thing?

LEE: I think it squarely is 2.0. The reason is, I don’t think of database virtualization as being incidental or pedestrian, kind of on the sideline and getting roped in with the wave. I think that database virtualization is one of the trends that leads the charge to Virtualizaton 2.0 and makes the adoption rate go up from 10 percent to something even more significant.

Why I say that is — and in speaking with all the hypervisor vendors I’m pretty sure they’re going to tell you the same thing — what it is that’s keeping that adoption rate going from 10 to 30 percent is that enterprise applications have to live on the hypervisor. This 10 percent was the easy adoption of taking development and test environments and turning them into really dynamic infrastructures — perfect sense, perfect economic payback, everything was clear.

To get to the next level of adoption, production environments with real enterprise applications that all access data have to be accommodated. You don’t get to Virtualization 2.0 — which we defined on the panel, loosely, as the acceptance of enterprise applications in a hypervisor environment — unless the database itself can be virtualized. Otherwise, you really under-power the whole platform. So it’s not a 3.0 play, it’s one of the drivers that will create the 2.0 experience.

ON-DEMAND ENTERPRISE: How big is the database virtualization market, in terms of both customers and vendors?

LEE: What we see all the time from a competitive perspective is that we bump up against database replication and database clustering 1.0, the tried-and-true legacy approach, whether it’s the database vendors’ mirroring or log shipping, the system vendors’ server clustering, or the storage vendors’ block-level replication. Whatever it is, I would consider those technologies to be the first wave of availability infrastructure, and what this new market is comprised of is the approach we’re taking.

It’s a pretty new market, and it’s one that catches a lot of IT people by surprise. I’m awfully sensitive to it, but all of those keynotes [at Next Generation Data Center] … talked about this brave new world and all the cool things they’re embracing, and everyone conveniently skipped over the database. I was sitting in a session by one of the big hosting companies, and they gave a great session on scaling, and yet when they talked about the database contribution to that scaling architecture, it was all replication and clustering 1.0 approaches — because people simply aren’t aware that technologies like Gridscale exist.

So I would say pure database virtualization technology is a new market, but the market that it addresses is a huge market that currently has its incumbents … and they’re the ones that our customers will tell you are being unseated because we’ve got something much better.

ON-DEMAND ENTERPRISE: Are there any particular vertical markets in which you’re seeing more uptake than in others?

LEE: While we are a technology play, and so we’re pretty neutral when it comes to the applications that would use us, we do see that certain vertical markets have a much lower pain threshold, and therefore need to have the continuous availability right out of the gates, as well as the scale. Those would be your regulated industries — financial services, health care — where we’ve got lots of anecdotes, such as one health care company where the board of directors came down on one IT executive and said, “Our business is at risk because of this use of 1.0 replication and clustering technologies, and you shall get off those, and you shall have 30 days to do so.” Those kinds of dramatic stories, I think, really underscore the need to have a non-stop data experience without incurring outages for applications when there is planned or unplanned downtime.

We’re [also] seeing a lot more retail in our customer mix, particularly with some pretty high-profile outages. Macys.com, I believe, last Christmas season had a fairly high-profile outage. Not only brand damage, but a real, tangible loss of revenue are pain points for retail. Anything that’s customer-facing, where the end-user can be quite fickle — “I can but object A, B or C from any number of sources, and if I have to endure some sort of outage, I’m already a click away from your competitors” — [is a potential use case for Gridscale].


ON-DEMAND ENTERPRISE: You’ve experienced substantial customer growth in the past year, and that’s due in part to your relationship with IBM. Can you talk a little about this growth and give a background of your relationship with IBM?

LEE: In the quarter we just closed, compared to the same quarter last year, we had a 400 percent increase in revenue. I’d say that the number of customers has increased significantly, as well. I really can’t pull a number out of thin air, but the revenue numbers speak for themselves. The traction we’re seeing, and the growth of adoption in multiple industries, has been pretty geometric. Why is that? Let me answer the second question.

Our relationship with IBM, from day one, has been a very strong one and it is increasing in value day by day. In March, we signed an agreement with them to have their DB2 client support organization, the same one that any customer would call into with a DB2 problem … handle level 1 and 2 support — worldwide, 24 by 7 — for Xkoto customers. We both recognize there is a lot stake, there is a lot of strategic value, and we want the customers to have a unified experience through the support channel and not have a no-man’s-land experience where vendors point to each other and drop the ball on support. That’s a very strong statement of the value of our partnership with IBM.

We continue to have a number of joint field opportunities and engagements — marketing, co-marketing, partner marketing, brand marketing — [and] we continue to have a number of development dialogues going on, right up to their VP of engineering. Everywhere you look in the IBM Software Group data practice or information management practice, we’ve got touch points, and it grows all the time.

I think it’s a very strong, win-win relationship. We each bring something very significant to the table, and you’re going to see more; there are going to be other announcements coming.


ON-DEMAND ENTERPRISE: You currently only support DB2, correct?

LEE: At the production grade, yes. In the next 30 days, you can expect to see an official release to support SQL Server.


ON-DEMAND ENTERPRISE: How do you see SQL support affecting market penetration?

LEE: I think it’s going to cause real acceleration because SQL Server is the fastest-growing database, and the whole convergence of Hyper-V, Windows Server 2008, refresh on SQL Server to ’08, the new Exchange server, etc., all coming out this year creates a lot of momentum and a lot of expectation. Particularly with the Windows Server 2008 release, the expectation is for Microsoft to take more and more of these big enterprise opportunities. To do so, one of the important functions the Gridscale technology is going to address for them is to provide this continuous availability and scale for SQL Server and allow it to compete with all the other big database vendors on the most mission-critical types of environments. We’re expecting to see a real strong increase in uptake in the SQL Server community.

We were at the SQL Server user event almost a year ago (and that was very early days for us) just more in the technology planning for the SQL Server release, and the people that came through to our booth pretty much had a singular need that they expressed, and they were all looking at those 1.0 replication technologies to address those needs. They’re the ones that we’re really reaching out to now, because they’ve got these very palpable pain points.


ON-DEMAND ENTERPRISE: Do you have partnership plans in place with Microsoft similar to what you’re doing with IBM?

LEE: Every partnership program is a little bit different. We are partnered with Microsoft — we’re part of their partner program, we’re part of their Empowered ISV program — [and] we enjoy very strong relationships, similar to those we enjoy with IBM, up and down the product organization, in terms of marketing, partnering and development, right up to the executive ranks. We also are developing very deep relationships with adjacent product groups — Hyper-V, System Center and so on.

It’s a little bit different from IBM in that it is earlier days, but we fully expect that we’re going to enjoy the same kind of very deep relationship that we have now with IBM — because we solve problems together.


ON-DEMAND ENTERPRISE: When you look into the crystal ball, what do you see for Gridscale and for the database virtualization market, in general? How will it fit into the predicted paradigm shift to cloud or utility computing?

LEE: I have a sense of what that would look like, very similar to any other technology, whether you talk about hardware or virtualization software. What I would expect to see is that as database virtualization becomes recognized as a solution to a real problem that’s out there, it’s going to become more and more intrinsic to the infrastructure. Just like the hypervisor and the VT types of approaches by the chip vendors to integrate certain software capabilities into the silicon, I expect that the database virtualization we have is going to get tighter and tighter into the infrastructure such that it will become the default behavior rather than an add-on appliance. You would see that in every regard, in terms of the installation, error reporting, alerting, configuration, management control. You’d see those all becoming seamless, and you wouldn’t know that Gridscale was actually a part of the infrastructure because it would become, more or less, baked in.

Just like the hypervisor, now, needs to mature and the value has to go up the chain — that’s why we see all kinds of investment around management, availability and disaster avoidance technologies for the hypervisor — we would expect to see the same thing for database virtualization. As the core functionality and the core delivery of continuous availability and scalability becomes baked into database infrastructure … it needs to move up into management, it needs to move up into those adjacencies to have more of a lateral benefit for the application that live on top of it. I would expect to see more coordination — you know, we talk about load balancing as one of the value propositions for spreading that query workload around — I would expect to see more of an enterprise transactional cooperation between the multiple tiers in an infrastructure. I would like to see coordinated transactions and have more of those both in more seamless, verticalized infrastructure up and down the stack, and a tighter integration with the applications that use the technology. It’s got to go up because when we’re successful, then Gridscale is going to be a de facto part of the data experience.


ON-DEMAND ENTERPRISE: Everything now seems to be about cloud computing or utility computing. How, if at all, does Gridscale fit in with this? Does it need to?

LEE: I don’t see a lot of change required to our existing product to support cloud computing. I think the big challenge for cloud computing is dynamic scale. … When it really gets interesting is when you have multi-tenanted infrastructures that can scale dynamically and be general-purpose — not simply having someone’s dedicated environment being hosted somewhere, but having something really general-purpose and very scalable in nature. To do that, the database becomes the constraint. You can do that with all different layers in the architecture or infrastructure, but databases being stateful, and therefore really anchoring or constraining just how far you can go, the database to be able to deliver that kind of massive, dynamic scale. I think Gridscale will prove to be an important enabling technology for relational databases in the cloud.

Subscribe to HPCwire's Weekly Update!

Be the most informed person in the room! Stay ahead of the tech trends with industry updates delivered to you every week!

Q&A with Nvidia’s Chief of DGX Systems on the DGX-GB200 Rack-scale System

March 27, 2024

Pictures of Nvidia's new flagship mega-server, the DGX GB200, on the GTC show floor got favorable reactions on social media for the sheer amount of computing power it brings to artificial intelligence.  Nvidia's DGX Read more…

Call for Participation in Workshop on Potential NSF CISE Quantum Initiative

March 26, 2024

Editor’s Note: Next month there will be a workshop to discuss what a quantum initiative led by NSF’s Computer, Information Science and Engineering (CISE) directorate could entail. The details are posted below in a Ca Read more…

Waseda U. Researchers Reports New Quantum Algorithm for Speeding Optimization

March 25, 2024

Optimization problems cover a wide range of applications and are often cited as good candidates for quantum computing. However, the execution time for constrained combinatorial optimization applications on quantum device Read more…

NVLink: Faster Interconnects and Switches to Help Relieve Data Bottlenecks

March 25, 2024

Nvidia’s new Blackwell architecture may have stolen the show this week at the GPU Technology Conference in San Jose, California. But an emerging bottleneck at the network layer threatens to make bigger and brawnier pro Read more…

Who is David Blackwell?

March 22, 2024

During GTC24, co-founder and president of NVIDIA Jensen Huang unveiled the Blackwell GPU. This GPU itself is heavily optimized for AI work, boasting 192GB of HBM3E memory as well as the the ability to train 1 trillion pa Read more…

Nvidia Appoints Andy Grant as EMEA Director of Supercomputing, Higher Education, and AI

March 22, 2024

Nvidia recently appointed Andy Grant as Director, Supercomputing, Higher Education, and AI for Europe, the Middle East, and Africa (EMEA). With over 25 years of high-performance computing (HPC) experience, Grant brings a Read more…

Q&A with Nvidia’s Chief of DGX Systems on the DGX-GB200 Rack-scale System

March 27, 2024

Pictures of Nvidia's new flagship mega-server, the DGX GB200, on the GTC show floor got favorable reactions on social media for the sheer amount of computing po Read more…

NVLink: Faster Interconnects and Switches to Help Relieve Data Bottlenecks

March 25, 2024

Nvidia’s new Blackwell architecture may have stolen the show this week at the GPU Technology Conference in San Jose, California. But an emerging bottleneck at Read more…

Who is David Blackwell?

March 22, 2024

During GTC24, co-founder and president of NVIDIA Jensen Huang unveiled the Blackwell GPU. This GPU itself is heavily optimized for AI work, boasting 192GB of HB Read more…

Nvidia Looks to Accelerate GenAI Adoption with NIM

March 19, 2024

Today at the GPU Technology Conference, Nvidia launched a new offering aimed at helping customers quickly deploy their generative AI applications in a secure, s Read more…

The Generative AI Future Is Now, Nvidia’s Huang Says

March 19, 2024

We are in the early days of a transformative shift in how business gets done thanks to the advent of generative AI, according to Nvidia CEO and cofounder Jensen Read more…

Nvidia’s New Blackwell GPU Can Train AI Models with Trillions of Parameters

March 18, 2024

Nvidia's latest and fastest GPU, codenamed Blackwell, is here and will underpin the company's AI plans this year. The chip offers performance improvements from Read more…

Nvidia Showcases Quantum Cloud, Expanding Quantum Portfolio at GTC24

March 18, 2024

Nvidia’s barrage of quantum news at GTC24 this week includes new products, signature collaborations, and a new Nvidia Quantum Cloud for quantum developers. Wh Read more…

Houston We Have a Solution: Addressing the HPC and Tech Talent Gap

March 15, 2024

Generations of Houstonian teachers, counselors, and parents have either worked in the aerospace industry or know people who do - the prospect of entering the fi Read more…

Alibaba Shuts Down its Quantum Computing Effort

November 30, 2023

In case you missed it, China’s e-commerce giant Alibaba has shut down its quantum computing research effort. It’s not entirely clear what drove the change. Read more…

Nvidia H100: Are 550,000 GPUs Enough for This Year?

August 17, 2023

The GPU Squeeze continues to place a premium on Nvidia H100 GPUs. In a recent Financial Times article, Nvidia reports that it expects to ship 550,000 of its lat Read more…

Shutterstock 1285747942

AMD’s Horsepower-packed MI300X GPU Beats Nvidia’s Upcoming H200

December 7, 2023

AMD and Nvidia are locked in an AI performance battle – much like the gaming GPU performance clash the companies have waged for decades. AMD has claimed it Read more…

DoD Takes a Long View of Quantum Computing

December 19, 2023

Given the large sums tied to expensive weapon systems – think $100-million-plus per F-35 fighter – it’s easy to forget the U.S. Department of Defense is a Read more…

Synopsys Eats Ansys: Does HPC Get Indigestion?

February 8, 2024

Recently, it was announced that Synopsys is buying HPC tool developer Ansys. Started in Pittsburgh, Pa., in 1970 as Swanson Analysis Systems, Inc. (SASI) by John Swanson (and eventually renamed), Ansys serves the CAE (Computer Aided Engineering)/multiphysics engineering simulation market. Read more…

Choosing the Right GPU for LLM Inference and Training

December 11, 2023

Accelerating the training and inference processes of deep learning models is crucial for unleashing their true potential and NVIDIA GPUs have emerged as a game- Read more…

Intel’s Server and PC Chip Development Will Blur After 2025

January 15, 2024

Intel's dealing with much more than chip rivals breathing down its neck; it is simultaneously integrating a bevy of new technologies such as chiplets, artificia Read more…

Baidu Exits Quantum, Closely Following Alibaba’s Earlier Move

January 5, 2024

Reuters reported this week that Baidu, China’s giant e-commerce and services provider, is exiting the quantum computing development arena. Reuters reported � Read more…

Leading Solution Providers

Contributors

Comparing NVIDIA A100 and NVIDIA L40S: Which GPU is Ideal for AI and Graphics-Intensive Workloads?

October 30, 2023

With long lead times for the NVIDIA H100 and A100 GPUs, many organizations are looking at the new NVIDIA L40S GPU, which it’s a new GPU optimized for AI and g Read more…

Shutterstock 1179408610

Google Addresses the Mysteries of Its Hypercomputer 

December 28, 2023

When Google launched its Hypercomputer earlier this month (December 2023), the first reaction was, "Say what?" It turns out that the Hypercomputer is Google's t Read more…

AMD MI3000A

How AMD May Get Across the CUDA Moat

October 5, 2023

When discussing GenAI, the term "GPU" almost always enters the conversation and the topic often moves toward performance and access. Interestingly, the word "GPU" is assumed to mean "Nvidia" products. (As an aside, the popular Nvidia hardware used in GenAI are not technically... Read more…

Shutterstock 1606064203

Meta’s Zuckerberg Puts Its AI Future in the Hands of 600,000 GPUs

January 25, 2024

In under two minutes, Meta's CEO, Mark Zuckerberg, laid out the company's AI plans, which included a plan to build an artificial intelligence system with the eq Read more…

Google Introduces ‘Hypercomputer’ to Its AI Infrastructure

December 11, 2023

Google ran out of monikers to describe its new AI system released on December 7. Supercomputer perhaps wasn't an apt description, so it settled on Hypercomputer Read more…

China Is All In on a RISC-V Future

January 8, 2024

The state of RISC-V in China was discussed in a recent report released by the Jamestown Foundation, a Washington, D.C.-based think tank. The report, entitled "E Read more…

Intel Won’t Have a Xeon Max Chip with New Emerald Rapids CPU

December 14, 2023

As expected, Intel officially announced its 5th generation Xeon server chips codenamed Emerald Rapids at an event in New York City, where the focus was really o Read more…

IBM Quantum Summit: Two New QPUs, Upgraded Qiskit, 10-year Roadmap and More

December 4, 2023

IBM kicks off its annual Quantum Summit today and will announce a broad range of advances including its much-anticipated 1121-qubit Condor QPU, a smaller 133-qu Read more…

  • arrow
  • Click Here for More Headlines
  • arrow
HPCwire