home About EPS Global Blog September 2019 The Open Compute Project Part 1

The Open Compute Project Part 1

Share on
Written by: Barry McGinley Avatar 9/26/2019

The OCP’s third regional summit took place on the 26th and 27th of September in Amsterdam. I was in attendance as EPS Global’s representative for the second time (I was on my best behaviour, honest). Over the next 2 blogs I will share some of the new technologies that were on show from the biggest names in the hardware world. In part 1 though, I will give an overview of the OCP, where, when and why they began, the ethos within and where it is heading. Part 2 will be dedicated to the new technologies that aim to solve the ever-increasing challenges within cloud computing and more. With members including Facebook, Google, IBM, Rackspace, Nokia, Microsoft, Alibaba Group and Cisco to name but a few, all working in collaboration, the Open Compute Project looks capable of rising to these challenges.
 

The History

The OCP’s story begins back in 2009 with Facebook. They began to grow exponentially, offering an ever-increasing range of services along with their platform for sharing what you had for your lunch. Looking ahead, they realized that in order to accommodate the massive influx of new people and data it would have to drastically rethink its infrastructure along with controlling running costs and energy consumption. Over the next 2 years a small group of engineers worked on building the worlds most efficient data center which could handle the unprecedented demands of Facebook's network.  This included the software, servers, racks, power supplies and cooling. After the completion of the data centre in Prineville, Oregon, the results were a DC 38% more energy efficient to build and 24% less expensive to run. This was just the beginning...

In 2011, Facebook decided to share the designs they had spent the previous 2 years working on with the public. Intel and Rackspace did likewise and along with the monster that is Goldman Sachs they founded the not for profit Open Compute Project. The idea behind the collaboration was to create a movement within the hardware industry that would bring about the same levels of creativity and partnership that is present in the open source software community. 8 years later and it is safe to say it has been a success.

OCP
 

How does it work?

It is easy to say you want to change the world when you are Facebook or Google but how do the smaller members within the OCP make a dent in the system? Well, once you are a member you can contribute to the community if your contribution hits 3 out of the 4 core tenets. It vigorously sticks to these tenets which breeds consistency and excellence in design and keeps the OCP as one of the leading organisations in the open source world. The 4 are as follows…

  • Efficiency – Examples of efficiency within a design are thermal efficiency, power delivery and conversion, weight, cost or latency reduction and platform performance.
  • Scalability – The technology must have the right supporting features to allow for maintenance in large scale deployments. Management tools need to be available from open sources and come with the appropriate documentation to allow for successful deployment.
  • Openness – All contributions must be open. In certain cases, 100% open source is not possible, but it is the goal. The contribution must comply with a set of already existing open interfaces, or at least provide one.
  • Impact – All contribution must make a meaningful positive impact in the OCP ecosystem. This can take place by building on top of already existing OCP solutions or improving the supply chain by providing compatible solutions. Other examples are utilizing new technologies or scale out technologies.

There are also different types of contributions to the OCP. There can be specifications without design files and then design files based on these approved specifications. These are used to create the products (I will come back to this), reference architectures, tested configurations, case studies and white papers. Let’s use Edgecore as an example here. Edgecore’s 5712- 54X 10GbE top of rack switch was the first switch accepted by the OCP and Edgecore have now contributed 29 different products. Once a product is accepted it enters the, ‘OCP Marketplace’.
 

OCP Marketplace

There would be little point or virtue to the OCP if it only benefitted the hyperscalers like Facebook or Google. The Marketplace has an online hardware catalogue of network gear, servers, storage, GPUs and rack & power. It also has a section on the software developers who have contributed to OCP like Microsoft’s SONiC, OcNOS from IPInfusion, Pluribus, PICOS from PICA8 and more.

The Marketplace allows smaller data centre operators and enterprises to access these products, vendors who sell the products and importantly, the specifications. Using Edgecore’s 5712 again as an example. Once you find the product through the network gear tab, or Edgecore tab, you can download all the design collateral. This includes a folder on hardware design, which incorporates every detail of the switch like BOM’s, fan boards, main boards, mechanical and CPU modules. There is also a folder on the software and specifications which leaves no stone unturned.
 

Don't miss out!

Get blogs like this straight to your inbox! Subscribe me to EPS's monthly Tech Roundup.

sign up

The future is bright!

2018 saw the OCP hardware ecosystem generate $2.56 billion in revenue which equated to a 120% growth rate when compared to 2017. IHS Markit expects this to surpass more than $10 billion by 2022.


Server, storage and networking equipment are the fastest projected growth areas with PON a potential high-growth area they said. Telco spend on OCP equipment is also expected to surpass hyperscaler spend by 2021. This is down to things like disaggregated cell site equipment, disaggregated routers and other work being carried out by the ONF (Open Networking Foundation) like CORD and SEBA. Part 2 of this blog will shed some more light on what the future will look like from the green tinted windows within the Open Compute Project.

As always, I would be more than happy to share additional resources with you or for more technical information on products or SDN give me a shout also you can browse our Open Networking products here.

Barry
 

Glossary of Terms

  • IoT – Internet of Things
  • 5G – 5th generation of cellular mobile communication
  • Linux – Family of free open-source operating systems
  • ONF – Open Networking Foundation
  • OCP – Open Compute Project
  • SDN – Software Defined Networking
  • Edgecore – White box ODM
  • Quanta – White box OEM
  • Data Plane – Deals with packet forwarding
  • Control Plane – Management interface for network configuration
  • ODM – Original design manufacturer
  • OEM – Original equipment manufacturer
  • Cumulus Linux – Open network operating system
  • Pluribus – White box OS that offers a controllerless SDN fabric
  • Pica8 – Open standards-based operating system
  • Big Switch Networks – Cloud and data centre networking company
  • IP Infusion – Whitebox network operating system 
  • OS – Operating system
  • White Box – Bare metal device that runs off merchant silicon
  • ASIC – Application-specific integrated circuit
  • CAPEX – Capital expenditure
  • OPEX – Operating expenditure
  • MAC - Media Access Control
  • Virtualization – To create a virtual version of something including hardware
  • Load Balancing – Efficient distribution of incoming network traffic to backend servers
  • Vendor Neutral - Standardized, non-proprietary approach along with unbiased business practices
  • CORD – Central Office Rearchitected as a Data Center
  • SD-WAN – Software Defined Wide Area Network
  • NFV – Network Function Virtualization
  • RTBrick – Web scale network OS
  • Snap Route – Cloud native network OS
  • MPLS – Multiprotocol label switching
  • DoS – Denial of service attack
  • ONOS – ONF controller platform
  • LF – Linux Foundation
  • MEC – Multi-access edge computing
  • Distributed Cloud -
  • COMAC – Converged Multi-Access and Core
  • SEBA – SDN enabled broadband access
  • TRELLIS – Spine and leaf switching fabric for central office
  • VOLTHA – Virtual OLT hardware abstraction
  • R-CORD- Residential CORD
  • M-CORD – Mobile CORD
  • E-CORD – Enterprise CORD
  • PON – Passive optical network
  • G.FAST – DSL protocol for local loops shorter than 500 metres
  • DOCSIS – Data over cable service interface specification
  • BGP – Border gateway patrol routing protocol
  • OSPF – Open shortest path first routing protocol
  • DSL – Digital subscriber line
  • Container – Isolated execution environment on a Linux host
  • Kubernetes – Open source container orchestration system
  • Docker – Program that performs operating-system-level virtualization
  • Cloud Native – Term used to describe container-based environments
  • CNCF – Cloud Native Computing Foundation
  • API – Application Programming Interface
  • REST API – Representational State Transfer Application Programming Interface
  • CLI – Command Line Interface
  • VM – Virtual machine
  • NAT – Network Address Translation
  • IBN – Intent Based Networking
  • TORs – Top of Rack Switches
  • RHI – Route Health Injections
  • BCF – Big Cloud Fabric
  • VPC – Virtual Private Cloud
  • ONIE – Open Networking Install Environment
  • CI/CD - Continuous Integration/Continuous Deployment





The Open Compute Project Part 1

The OCP’s third regional summit took place on the 26th and 27th of September in Amsterdam. I was in attendance as EPS Global’s representative for the second time (I was on my best behaviour, honest). Over the next 2 blogs I will share some of the new technologies that were on show from the biggest names in the hardware world. In part 1 though, I will give an overview of the OCP, where, when and why they began, the ethos within and where it is heading. Part 2 will be dedicated to the new technologies that aim to solve the ever-increasing challenges within cloud computing and more. With members including Facebook, Google, IBM, Rackspace, Nokia, Microsoft, Alibaba Group and Cisco to name but a few, all working in collaboration, the Open Compute Project looks capable of rising to these challenges.

The OCP’s third regional summit took place on the 26th and 27th of September in Amsterdam. I was in attendance as EPS Global’s representative for the second time (I was on my best behaviour, honest). Over the next 2 blogs I will share some of the new technologies that were on show from the biggest names in the hardware world. In part 1 though, I will give an overview of the OCP, where, when and why they began, the ethos within and where it is heading. Part 2 will be dedicated to the new technologies that aim to solve the ever-increasing challenges within cloud computing and more. With members including Facebook, Google, IBM, Rackspace, Nokia, Microsoft, Alibaba Group and Cisco to name but a few, all working in collaboration, the Open Compute Project looks capable of rising to these challenges.


EPS Global
Comments

Blog post currently doesn't have any comments.

Need Help?

We have local language and currency support in each of our 28 locations, ensuring you always have access to friendly customer support to deliver your hardware solutions regardless of your location.

Request a callback from our experts