Products > Product Portfolio > vSMP Foundation for Cloud
vSMP Foundation for Cloud
Cloud computing describes an Internet based computing infrastructure that has abstracted users and users applications from the underlying computing resources that support them. In concept this is functionally different from previous IT architectures in that users no longer need to own, have expertise in or control over the underlying technology and are only aware of borrowing and consuming IT services much as they would telephony, electrical or plumbing infrastructure. This Cloud paradigm has developed along with the maturity of virtualization and provisioning technologies that enable resources-as-a-service using internet protocols.
From Wikipedia:
"A technical definition is "a computing capability that provides an abstraction between the computing resource and its underlying technical architecture (e.g., servers, storage, networks), enabling convenient, on-demand network access to a shared pool of configurable computing resources that can be rapidly provisioned and released with minimal management effort or service provider interaction." This definition states that clouds have five essential characteristics: on-demand self-service, broad network access, resource pooling, rapid elasticity, and measured service."
The challenge for evolving Cloud Computing architectures however is that there have been and will always be significant workloads that can't easily be virtualized using existing compute nodes (virtualized inside a single box). These applications, often segmented as high-performance workloads, require either a large number of processing cores or large memory / RAM footprints that are beyond the scale of normal Cloud Computing resources. To end users, who have increasingly been disaggregated from the compute nodes and trained to think of compute resources as on-demand, this is both frustrating and challenging. They have to differentiate workloads as those that can run in the cloud and those that they need to continue to buy dedicated, expensive and purpose built hardware. These users also need to maintain and manage these additional compute resources when their energies are likely better spent running workloads and generating applications results instead of managing IT.
With ScaleMP's vSMP Foundation for Cloud™, users utilizing Cloud Infrastructure no longer have to differentiate workloads as IT can now carve out large SMP systems -- up to 128 cores and 4TB of RAM -- out of smaller nodes in their existing x86 infrastructure. This will allow IT administrators to modify their compute resources to the needed workload rather than restricting the infrastructure to the workloads that fit within a single node. This increase in workload addressability will, in most cases, increase cloud utilization to close to 100% of enterprise and compute workloads and increase Cloud infrastructure ROI!
vSMP Foundation for Cloud:
A New Way to Optimize x86 Resources The vSMP Foundation for Cloud solution enables on demand virtual SMP provisioning for enterprise cloud infrastructure. Tightly integrated with your existing systems management and provisioning tools, vSMP Foundation for Cloud offers the flexibility of on-the-fly creation of virtual machines on a per job, per project, or per customer basis, eliminating the need to deploy dedicated shared memory systems.
Using vSMP Foundation for Cloud allows for real-time deployment of large memory systems aggregated from cloud resources and redeployment back to the compute grid upon completion of application or batch jobs. Similar to the rest of the vSMP Foundation product lines, it supports aggregation of up to 16 servers, utilizing the latest generation of Intel Xeon processors (Nehalem - 55xx series) into a single virtual machine, with up to 128 cores and 4TB or main memory.
Unlike the other vSMP solutions, vSMP Foundation for Cloud is delivered in a network bootable image for easy Cloud installation. In addition, the initial software license for vSMP Foundation for Cloud is priced higher than the static solutions but significantly lower on subsequent per node licenses.
For static or dedicated system resources see vSMP Foundation for SMP.
Managing Cloud Aggregation
There are quite a few tools available for managing x86 server resources via a network. The most important of these tools include workload modeling, scheduling and provisioning. While ScaleMP has tested most of these tools -- particularly the provisioning solutions from Adaptive Computing and IBM -- since the vSMP Foundation for Cloud software is built on top of standard APIs specifically for network provisioning the solution integrates well with all existing solutions including those from Platform Computing, Univa, HP and others.
By using these tools Cloud infrastructure managers can identify the available resources for the workload, schedule those resources for aggregation and provision both vSMP Foundation for Cloud to aggregate and then install both an OS and the applications necessary to run the workload. Upon completion the provisioning software can disaggregate the nodes and reallocate those resources back into the Cloud pool.
For more on the full modeling, scheduling and provisioning integration with Adaptive Computing be sure to take a look at the joint statement of support announcement.
This SMP on demand solution from ScaleMP is so robust that there are customers today not only running vSMP Foundation for Cloud for their internal cloud deployments but also customers in the small but growing HPC-as-a-service space using ScaleMP's on the fly aggregation technology to create external service provisioned solutions. R Systems in El Dorado Hills, CA is an example of one such cutting edge service provider.
Key messages:
- Dynamic, on the fly provisioning of systems to create large CPU/memory systems aggregation;
- Provisioning and reprovisioning of resources using existing infrastructure;
- Significantly extend the flexibility and utility of your existing Cloud investments Which vSMP Product is right for you?