Cray’s president and chief executive officer, Peter Ungaro, offers insights into the company’s new supercomputing-as-a-service offering in partnership with Markley. The first industry offering is the Cray Urika-GX agile analytics platform for life sciences organizations.
Q: Pete, tell us about this announcement.
Ungaro: One of the things that customers have been asking us for is the ability to get access to a Cray supercomputer in different ways other than purchasing one, putting it into their data center, and managing it themselves, as not all customers have the facility, staff, or budget. Some want the option to use OpEx dollars and to not have to host and manage the system themselves. Our first example of this was our partnership with Deloitte, in which we have a Cray supercomputer on the Deloitte cloud and Deloitte offers cyber security services to customers with it. This announcement is a further step in this direction as we are partnering with Markley to provide true supercomputing-as-a-service. This offering is really the first of its kind in the industry, leveraging a real supercomputer in the cloud and specifically targeting the life sciences industry.
There’s a clear distinction between supercomputing in the cloud and what is commonly today called “HPC in the cloud.”
Q: Is this truly a “first” if other organizations offer HPC in the cloud?
Ungaro: There’s a clear distinction between supercomputing in the cloud and what is commonly today called “HPC in the cloud.” Most all the larger cloud providers have something they call “HPC in the cloud,” which is typically a commodity cluster using either Ethernet or InfiniBand networks. While this is great for some customers, it is not meeting all customers’ needs, especially those customers after higher performance. What Cray and Markley are bringing to the table is our unique Urika-GX analytics system with the Cray Aries interconnect and Cray tools and libraries — basically, the whole Cray environment. Every research scientist, data scientist, or line-of-business owner who wants to run high-end analytics can now have the ability to access a Cray system. I think that this should be a very compelling and price-performant offering for a wide variety of customers.
Q: Why did Cray choose Markley?
Ungaro: Markley is a great company and a premier multi-tenant data center and hosting facility on the east coast. They have an incredible pedigree and they think of customers first, just like Cray. Markley is also well-known in the life sciences research community, bringing those customers targeted and flexible computing solutions. Together, we thought we could offer something unique that would address our respective clients’ needs and not require Cray to build and run our own cloud service, which is not a core competency of ours but, of course, is of Markley’s. We each get to focus on what we do best and bring that together for customers.
Q: Why start with the life sciences industry?
Ungaro: We’re starting with life sciences because Markley has a well-established reputation of excellence within the life sciences community for hosting infrastructure and because the Cray Urika-GX system has demonstrated incredible capability in next-generation sequencing with the incredible growth of data that customers in this industry are seeing. In our work with life sciences organizations, we’ve demonstrated our ability to positively affect analytics workflows, providing up to 5x speedups for aspects of genomics variant analysis, for example. We’ve also been able to speed software development cycles, especially with the big data sets common to the life sciences. This results in more ability to explore their data, and better software tools to do the exploring.
…there are more and more reasons today for research scientists, data scientists, and enterprise institutions of all types to have access to the best tool available to get their work done.
Of course, this offering is available to all customers, not just those in the life sciences industry, as there are more and more reasons today for research scientists, data scientists, and enterprise institutions of all types to have access to the best tool available to get their work done. Many are looking for something that has more capability than what they can get today, the kind of performance and capability that Cray uniquely provides. This new service makes it easier for them to access the best tool available — a Cray supercomputer — in partnership with Markley, who can provide the service levels and infrastructure they need.
Q: Why would a user choose Cray as a service rather than invest in their own systems or use something like AWS?
Ungaro: For many companies and institutions, having a Cray on site makes the most sense. It offers them the most capability for the least amount of money and is a big TCO win for them. For others, having it as a service fits their institutional requirements and directions, their budget requirements, their computing needs. Some do a little of both. It’s not an “or,” it’s an “and.”
Why not AWS? This goes back to my earlier point — there’s a difference between commodity HPC clusters and supercomputing capability. Supercomputing capability is about faster time to insight, faster time to decisions. Today, large cloud companies have HPC cluster offerings that are really about capacity computing — “I can get you a whole bunch of compute cores, but it will likely take longer to run your application.” This offering provides customers a cost-effective way to get exceptional performance, at an attractive price, so they can have the best of both performance and capability as well as TCO.
Q: Can you tell us when Cray will expand the supercomputing-as-a-service offering to include more products besides the Urika-GX data analytics platform?
Ungaro: As I mentioned earlier, we want to continue to expand the ways customers can have access to Cray systems. So, all I’ll say right now is “stay tuned!”