Intel® Rack Scale Design Is Now Ready for Open Source Development

Today marks an important milestone for Intel® Rack Scale Design, the first industry-standard framework for disaggregating and dynamically managing compute, storage, and network resources for more efficient utilization of data center assets. We mark the efforts of our ecosystem partners as Intel® Rack Scale Architecture transitions to Intel Rack Scale Design and goes live as open source code.

As a VP in Intel’s Data Center Group and GM of the newly formed Intel Scalable Datacenter Solutions Group, I’m particularly excited about this milestone. I’m responsible for creating full solutions necessary to meet the most critical infrastructure challenges across the data center and working with the industry to bring additional ecosystem-led innovations to those solutions. I previously led Intel’s High Performance Computing Platform Group, where we established Intel® Scalable System Framework—a flexible, systems-level approach to delivering complete solutions.

Today, my focus is on the great progress we have made over the past year with Intel Rack Scale Design for hyper-efficient data centers. The release of the open source code for Intel Rack Scale Design heralds a significant step toward the revolution of traditional rack servers for organizations that view the data center as a strategic asset. The updated name for the framework reflects the longer-term vision of agile infrastructure that is at the heart of Intel Rack Scale Design.

A year ago at the Intel Developer Forum (IDF 2015), Intel hosted an Intel Rack Scale Architecture Developers Conference. Several OxMs—including Dell, Quanta Cloud Technology (QCT), Ericsson, and Inspur—took to the podium to speak to an enthusiastic crowd about the opportunities Rack Scale would open to them and their customers.

Intel Rack Scale Design is the first framework to be based upon and use the Redfish™ industry standard from DMTF for modern and secure management of scalable platform hardware in the modern data center. The framework allows for dynamic management of compute, memory, PCIe, and storage resources and the pooling of those resources for more efficient use of data center assets. The framework simplifies advanced technology to accelerate the adoption of open, interoperable solutions for tomorrow’s data centers today.

This is the first step in preparing the broader ecosystem for pooled resources and a path to a software-defined infrastructure.  Through its ability to provide a new systems-level architecture that uncouples a system’s resources, Intel Rack Scale Design helps hyperscale operators address the challenges of growing workload complexities and the sheer scale of usage demands.

The industry is taking note of that point. Craig LeGrande, managing director of Mainstay Salire LLC and author of “An Economic Study of the Hyperscale Data Center,” says that “Intel’s RSD framework will re-write the book on (hyperscale) data center operations. The ability to simplify and scale data center infrastructure through disaggregation and dynamic management of compute, storage, and network resources will provide strategic advantages to businesses that require high performance, nimble IT systems while reducing overall data center TCO.”

LeGrande notes that Mainstay’s research found that data centers powered by Intel Rack Scale Design infrastructure “can reach up to four times the level of CPU utilization and ten times the level of IT admin productivity compared to traditional data center environments—two of the largest drivers of data center TCO.”

In a notable step forward, Intel is now making the Intel Rack Scale Design components available as open source software that developers can productize and incorporate into orchestration solutions. With this open source release, Intel provides a clear path to address data center management fragmentation, drive down total cost of ownership via resource pooling, and improve OpEx via enriched telemetry.

Intel Rack Scale Design is coming to fruition because of the efforts of an ecosystem of partners who worked together in the spirit of open source collaboration. The specifications as well as all of the APIs associated with Rack Scale are open and freely available. The Intel Rack Scale Design APIs are based on the Redfish standard developed by DMTF. Intel Rack Scale Design is under the umbrella of the Intel® Cloud for All initiative which is focused on unleashing tens of thousands of new clouds.

You can now take advantage of the core of the Intel Rack Scale Design components. Currently available components include the framework’s Pod Manager, Pooled System Manageability Engine, Rack Management Module, and Validation Test Suite—a tool to aid in verifying that the implementation is Rack Scale aligned.

Today, the release of the code to the open source community is the first step forward in the widespread effort to enable the highly efficient use of resources. It sets up the hardware reference and open source APIs to simplify management, and it also sets up the ecosystem for pooled high-speed storage and future pooled technologies.

We expect solutions based on Intel Rack Scale Design to be available from multiple hardware and software vendors by the end of this year. To date numerous nodes have been deployed at multiple telco and cloud service providers via our hardware and software partners. And, yes, as with any truly revolutionary idea, we recognize that the shift to a new hyperscale design framework will take time. We also know that now is the time to start deploying the next generation of hyper-efficient data centers.

You will have a chance to see for yourself at this year’s IDF16 conference in San Francisco, which will showcase Intel Rack Scale Design in demos by Intel and our partners, as well as demos in multiple tech sessions.

In addition, you can download the code now via GitHub (https://github.com/01org/intelRSD) to begin exploring how you will innovate with Intel Rack Scale Design. Several designs across multiple markets and geographies from major OEMs and TEMs will be available in the marketplace by the end of the year.

For a closer look at this new framework for the data center, visit http://www.intel.com/IntelRSD.

I will close with a list of our partners for solutions based on Intel Rack Scale Design.

American Megatrends Inc. (AMI): https://ami.com/intel-rsd-pr/ https://ami.com/intel-rsd/
Dell: http://Del.ly/D13s49
Ericsson: http://cloudblog.ericsson.com/intel-rack-scale-design
HPE: http://www8.hp.com/us/en/products/servers/scalable-systems/cloudline/index.html
Inspur: http://www.inspursystems.com/products/incloud-rack/
Lenovo: http://blog.lenovo.com/blog/entry/2277
Quanta Cloud Technology (QCT): http://www.qct.io/
Super Micro Computer Inc.: https://www.supermicro.com/solutions/SRSD.cfm

 

Published on Categories Uncategorized
Charles Wuischpard

About Charles Wuischpard

Charles Wuischpard is vice president of the Data Center Group and general manager of the Scalable Datacenter Solutions Group at Intel Corporation. In his role, Wuischpard leads delivery and execution of datacenter solutions supporting the Enterprise, High Performance Computing, Cloud and Communications segments. Prior to joining Intel in 2013, Wuischpard was CEO and president of Penguin Computing, repositioning the company into emerging data center markets and launching cloud solutions targeted at the high performance computing segment which resulted in profitable growth of over 40% per year over 5 years. Prior to his role at Penguin, he held vice president positions at Golden Gate Software and Versant, and nearly 17 years at IBM in a career that ranged from chip designer to corporate sales and marketing. Wuischpard received a BSEE in Microelectronics from Purdue University and a post-graduate concentration in finance and international business from Northwestern University’s Kellogg School of Management.