Following EEMBC’s long-standing tradition of developing reliable, credible, and equitable benchmarks, we are excited to announce the formation of our Cloud and Big Data Benchmark Working Group. Collaboratively, we will provide an industry standard suite of performance and efficiency benchmarks that address the needs of device (ODMs) and equipment (OEMs) manufacturers providing compute systems to the scale-out datacenter marketplace and their consumers.
Addressing the Needs of Cloud and Big Data
Big Data is huge – literally, in terms of the petabytes of data that must be processed, and financially, in terms of the tremendous growth that is ongoing in the distributed computing environment. The operators of Hyperscale data centers are requiring faster and more power-efficient distributed computing platforms, being driven by the need to provide increasingly scalable online services and to handle the enormous demand generated by Internet of Things applications. To help quantify the performance and energy efficiency of computing platforms, the industry needs professional-grade benchmarks that could ultimately lead to the acceleration of a wide variety of real-world applications, including those used in unstructured data stores, web caching, web serving, elastic search, media streaming, data analytics, distributed cloud storage, and graph analytics.
Cloud and Big Data Benchmark Working Group Activity
With development occurring in phases, this EEMBC working group will strategically address a broad range of application categories commonly found in datacenters. The first phase will include real-world workloads represented by graph analytics, media streaming, and data caching.
Key Characteristics of the EEMBC Cloud and Big Data Benchmark include:
- Automated install and build process to ensures consistent execution (multiplatform support)
- Relatively low cost to implement (without requiring a large infrastructure)
- Predictable performance at scale to account for scale-out compute clusters, networking fabrics, and distributed storage and memory
- Repeatable, verifiable, and certifiable - as in other EEMBC benchmarks
Working Group Status
- Currently developing the first phase workloads
- Join the EEMBC Cloud and Big Data Server working group to help ensure a meaningful and fair representation for your company’s products. Email EEMBC for more information.
- Shay Gal-On, Cavium, Inc.
"EEMBC's Cloud and Big Data Server Benchmark working group is aimed at filling a significant gap for microserver and hyperscale infrastructure vendors and customers - creating a set of processor and system architecture neutral benchmarks for high value workloads." Paul Teich,
TIRIAS Research, Principal Analyst.
“Equipment manufacturers, and cloud and big-data users, have an unprecedented choice of solutions available for the distributed data centers with clusters of servers required for Cloud computing and Big Data. Historic benchmarks have proved inadequate for quantifying and comparing different server and processor architectures when executing these types of workload. The benchmarks being developed by the EEMBC cover the whole system including processor, memory, storage, middleware, and I/O, and will enable equipment manufacturers, and users to significantly improve the evaluation and planning of server solutions for real-world cloud and big-data applications.” Simon Stanley, Founder and Principal Consultant, Earlswood Marketing Limited, Contributing Analyst, Heavy Reading.
“In our analysis of processors and systems targeting the cloud and big data applications, we found a lack of benchmarks that are representative of real-world cloud workloads. We are excited to see that EEMBC is tackling this problem with their new Cloud and Big Data working group initiative and encourage cloud system hosts and cloud system developers to get involved in this effort to help EEMBC deliver fair and equitable benchmarks for this growing market segment.” Linley Gwennap, Principal Analyst of The Linley Group.