Using Intel® RSD to Reduce Data Center Power Consumption

Author Published on

Version 2.2 of Intel® Rack Scale Design Released—Supports Purley Advanced Telemetry and TPM

Intel® Rack Scale Design (Intel® RSD) 2.2 supports the Intel® Xeon® Scalable Family platform with out-of-band (OOB) discovery of features such as Trusted Platform Module (TPM), advanced telemetry and FPGA accelerator cards. In addition, OOB telemetry support provides detailed information about the health and status of infrastructure components to aid in provisioning and for capacity, performance, power, and thermal management.

The agentless OOB implementation continuously gathers telemetry data without impacting the performance of running applications, and also improves security by isolating management controls for the data network. Version 2.2 auto-discovers PCIe add-in FPGA accelerator cards to allow composition of nodes with FPGAs, and enabling orchestration software to assign FPGA resources to specific workloads. TPM support includes the ability to discover, enable, disable and clear TPMs. The latest reference documents are available on the Intel® RSD web site and the reference code is on Intel® RSD GitHub.

Intel® RSD Telemetry Helps Reduce DC Energy Costs

Customers are already using the new telemetry capabilities of Intel® RSD 2.2 to enable advanced data center solutions. For example, cloud service provider (CSP) SafeDX* is using Intel® RSD to implement improved temperature controls to reduce energy consumption.

With cooling equipment consuming 40% of the total data center power budget, cloud service providers (CSPs) and enterprises are looking to reduce cost by improving power usage efficiency (PUE). A growing trend is to save power by operating data centers at higher ambient temperatures. Employing this technique successfully requires precise monitoring of equipment to keep local temperatures within safe maximums.

The SafeDX solution uses:

  • Telemetry capabilities built into the Foxconn R-5211* server based on the Intel® Xeon® Scalable processor platform.
  • Intel® RSD with its support for OOB telemetry gathering and communication via open APIs.
  • Intel® Intelligent Data Center Management (Intel® IDCM) software.

Together, these technologies provide precise, local thermal and power monitoring of racks and individual servers, enabling an effective feedback and control loop to improve data center cooling efficiency. SafeDX* has been able to demonstrate a 17% cooling energy reduction using this closed loop approach, and is on track to deliver a PUE of 1.5 in the next deployment.

The SafeDX energy efficiency architecture shown above employs temperature sensing at the device level, Intel RSD software components and APIs at the control level to access and relay the sensor data, and IDCM software at the management layer.
SafeDX Energy Efficiency

The SafeDX* energy efficiency architecture shown above employs temperature sensing at the device level, Intel® RSD software components and APIs at the control level to access and relay the sensor data, and IDCM software at the management layer. Working together, Intel® RSD software components (PSME, RMM and PODM) collect telemetry data such as fan speed, airflow, CPU thermal margin, DIMM thermal margin, inlet and outlet temperatures, CPU/memory/ system utilization, and power consumption data from each rack and server node.

PODM relays all this information to IDCM, which maintains overall control of the cooling system. IDCM provides backend services including a sampler service, Thermal Control Strategy (TCS) service, and a database. The sampler service leverages PODM APIs to gather thermal and power telemetry data (servers and rack infrastructure) and stores it in the IDCM database. The TCS service analyzes the data and calculates efficiency metrics such as return temperature index (RTI) and rack cooling index (RCI). Based on its analysis, IDCM sends commands to the air conditioning units to adjust thermal settings. Real-time data and metrics are also presented to operators through a GUI dashboard.

We expect to see much more data center innovation in the future based on the new hyperscale monitoring and control capabilities provided by Intel® RSD via open, modern RESTful APIs. Read mode about the Foxconn/SafeDX energy efficiency solution here. Also check out the related video.

More on Intel® Rack Scale Design

To learn more about Intel® RSD or get more involved in the Intel® RSD ecosystem, check out the Intel® RSD website. Or contact your OEM or Intel sales representative now for an evaluation of Intel® RSD based systems. If you are planning to attend the OCP Summit*, please visit Intel and our partners to see Intel® RSD for yourself.

Be sure to follow me on Twitter @UlgenFigen (https://twitter.com/UlgenFigen).

Published on Author Categories Cloud ComputingTags , , , , ,
Figen Ulgen

About Figen Ulgen

Figen Ülgen, Ph.D., is General Manager of the Rack Scale Design Group within the Datacenter Group at Intel Corporation. Ülgen oversees solutions roadmap, solutions enablement as well as operations, support and ecosystem and partnership strategy for Rack Scale Design. Ülgen joined Intel in 2014, where she led as General Manager for HPC Software and Cloud developing and promoting HPC Platform Software technologies for the HPC ecosystem. She founded the award-winning OpenHPC Community under Linux Foundation with a group of ecosystem partners to democratize HPC. During her time with the HPC Software and Cloud, the team shipped Intel® HPC Orchestrator, a fully supported production quality software product in less than 18 months—concept to General Availability. Prior to that, she was a Senior Director with Microsoft for over a decade in various leadership roles in Data Science, Azure Public Cloud, and HPC teams. Her 20-year industry experience includes technology innovation, competitive strategy, software product planning and management, as well as driving change across large organizations. Before Microsoft, Ülgen worked at McKinsey & Co., Motorola and at Justsystem Corporation in Japan. With a passion for serving underprivileged youth and women in the technology field, Ülgen sits on the Saturday Academy Board of Directors and is regularly invited to speak on behalf of women in technology. Ülgen earned a Bachelor’s Degree in Computer Engineering, a Master’s Degree in Expert Systems under a Fulbright Scholarship and a Ph.D. in Machine Learning under the Japanese Government Ministry of Education Scholarship. Ülgen holds two U.S. patents.