Grid Virtualization and Itanium®-based Solutions
NextAxiom®, a member of the Itanium® Solutions Alliance, has taken this concept to the next level by leveraging the power of Intel® Itanium® processor solutions with what are now broadly known as Virtualized Service Grids.
Great. But how does this work? In a 2007 case study, NextAxiom's founder and CEO Ash Massoudi noted that the idea goes beyond the concept of service oriented architecture (SOA) and the use of standard web-based service systems, which commonly rely on service operations to communicate from system to system. Instead, NextAxiom replaces this with what they call "Service Oriented Programming," or SOP. Using SOP, the key concept is to create an "in-memory" image of services that run an entire application transparently without an IT department having to deploy a complex web of SOA middleware or load-balancing solutions to make the optimum use of the virtualized grid. The application runs on a Service Virtual Machine, then automatically (through NextAxiom's Hyperservice business platform) layers it across the virtual grid, allocates resources, and distributes the computing load transparently to you, the user.
It's a perfect match for an Itanium-based solution, because, as Massoudi noted in an interview last year, "multi-core processor architectures like the Dual-Core Intel Itanium processor provide the most cost-effective and efficient foundation for Virtualized Service Grids". Add to that the built-in virtualization features of Intel Itanium processors, with its raw computing power and scalability, and you have a virtualization solution unmatched for this kind of application.
How does it work in practice? As noted in a case study available on this site, NextAxiom's Hyperservice Studio allows users to simply drag and drop services, add visual programming elements, and then link them together. The software then automatically creates the service program from this design and applies it over the virtualized grid as needed, with little to no intervention from the application developer. Where feasible, programs are automatically parallelized and distributed across multiple platforms as needed. New services can be added or removed on-demand.
One major user of this solution is Ventyxtm, whose Service Delivery Management Solution (SDM) uses NextAxiom's SOA-based platform, running on a parallel HPC system using HP Integrity rx6600 servers, powered by four Dual-Core Intel® Itanium 2 processors and 96 GB of RAM.
How do you know if these new SO-based platforms are what you're looking for? If you're running service-oriented solutions with high-volume needs, you should consider this option. Grid computing and virtualized service applications have definitely come into their own.
All views expressed within this blog are those of the blog author and do not reflect the opinions of the Itanium Solutions Alliance, Itanium Solutions Alliance member companies or Intel Corporation.
