Swipe to navigate through the chapters of this book
Cloud operators realized early on that servers originally designed for enterprise data centers were not optimal for their application. These servers are well known branded machines from manufacturers such as Dell, Hewlett Packard Enterprise or SuperMicro, known as original equipment manufacturers, or OEMs. Instead, they deploy in house designs with machines built by contract original design manufacturers, or ODMs. The cloud platform ecosystem represents an evolution of the standard high volume enterprise server ecosytem that started in the early 1990s. We document this evolution and look at some of the business and technology opportunities that this evolution brings.
Please log in to get access to this content
To get access to this content you need the following product:
This capability is also known in the industry as reliability, availability, and serviceability (RAS).
C. Gordon Bell, C. Mudge, and J. E. McNamara, Computer Engineering: A DEC View of Hardware Systems Design (Bedford, MA: Digital Press, 1978).
C. Gordon Bell, “United Engineering Foundation and IEEE STARS Program and Engineering and Technology History Wiki,” http://ethw.org/Rise_and_Fall_of_Minicomputers , February 2016.
BIOS stands for basic input/output system, a preboot environment that runs immediately after a platform is powered on implementing low-level I/O and configuration functions. For most platforms, a third-party independent software vendor develops the BIOS.
BMC stands for baseboard management controller, a microcontroller installed in a baseboard to implement platform management functions such as power and thermal management, chassis intrusion, and orchestration of the functions of various sensors and actuators installed in the baseboard. In more advanced servers the BIOS implements complex runtime functions such as adding processors to the mix and reconfiguring the system after a memory failure. Most OEMs and ODMs outsource the BMC firmware to third-party firmware independent software vendors (ISVs).
Redfish is a DMTF standard specification and schema using a RESTful interface using JSON to describe the data object and allows rapid integration of server management. See https://www.dmtf.org/standards/redfish .
Moor Insights & Strategy, “Quanta’s Server Business: Can They Scale Beyond Hyperscale?,” http://www.moorinsightsstrategy.com/research-paper-quanta%E2%80%99s-server-business-can-they-scale-beyond-hyperscale/ 2014.
P. Moorhead, “OCP Summit: HP and Foxconn Blur Server Industry Lines with Cloudline,” Forbes, http://www.forbes.com/sites/patrickmoorhead/2015/03/10/ocp-summit-hewlett-packard-and-foxconn-blur-server-industry-lines-with-cloudline/#41f22b733e94 , March 10, 2015.
See https://aws.amazon.com/products/ for a current listing.
Data Center Energy Usage Report (Lawrence Berkeley National Laboratory: U.S. Department of Energy, 2016).
A. Efrati, “Inside Uber’s Engineering Struggles ,” The Information, https://www.theinformation.com/inside-ubers-engineering-struggles , September 21, 2015; R. Miller, “Uber Scales up Its Data Centers to Support Growth, Data Center Frontier,” http://datacenterfrontier.com/uber-data-center-expansion/ , January 12, 2016.
Y. Izrailevsky, S. Vlaovic, and R. Meshenberg, “Completing the Netflix Cloud Migration ,” https://media.netflix.com/en/company-blog/completing-the-netflix-cloud-migration , February 11, 2016; J. Brodkin, “Netflix Finishes Its Massive Migration to the Amazon Cloud,” Ars Technica, http://arstechnica.com/information-technology/2016/02/netflix-finishes-its-massive-migration-to-the-amazon-cloud/ , February 11, 2016.
- Evolution of Cloud Server Platforms
- Sequence number
- Chapter number
- Chapter 4