What is Open Compute Project(OCP)

 Open compute project is an open source hardware initiative for data center environment. In the past, you might have heard about so many open source software, operating systems(Linux, Unix), word press applications, etc. When considering the best benefit of open source, which allows anybody to obtain the source code and modify the software which is of their own vision and mission without a vendor specific lock is a fantastic idea. You will have absolutely brilliant control of the complete product. But have you noticed that all of these meant to be for software related and none this was not available for hardware? This was the main idea behind the initiative of open compute project(OPC) which is proposed by Facebook in 2011.

The Open Compute Project (OCP) is reimagining hardware, making it more efficient, flexible, and scalable. The promise was the flexibility of hardware and software and designs for greater power efficiency and cost efficiency. OPC was initiated by Facebook and very quickly Intel, Rackspace, Goldman Sachs and Sun Microsystems' co-founder Andy Bechtolsheim joined with Facebook to launch the OCP project, with Microsoft joining in 2014.

Facebook hoped that other companies would adopt and adapt its initial designs, pushing down costs and improving quality – and they have: Sales of hardware built to Open Compute Project designs topped $1.2 billion in 2017, double the previous year, and are expected to reach $6 billion by 2021. Those figures, from IHS Markit, exclude hardware spending by OCP board members Facebook, Intel, Rackspace, Microsoft and Goldman Sachs, which all use OCP to some degree. The spend is still a small part of the overall market for data-center systems, which Gartner estimated was worth $178 billion in 2017, but IHS expects OCP’s part to grow 59 percent annually, while Gartner forecasts that the overall market will stagnate, at least through 2019.

Note: Even though OCP is mainly concentrated on hardware. The customizations are often in the software too(Like firmware or OS which controls these devices).

How did they achieve this?

When Facebook designed the hardware for its first dedicated data center in Prineville, Oregon, it wanted to make savings on three fronts: energy, materials, and money.

It boosted energy efficiency by cutting wastage in the power supply and by making the servers taller, which left room for bigger, more effective heatsinks and meant that it could use fans of a larger diameter, able to move more air with less energy. By doing away with vanity faceplates, paint, logos, unneeded expansion slots and components such as video cards and even mounting screws, it saved more than 6 pounds of material per server.

That inevitably led to cost reductions, as you don’t pay for the electricity you don’t consume or parts you don’t use. On top of that, it made savings on labor: Without the mounting screws, racking and unracking servers were quicker; standardization saved time dealing with spares, and overall systems could be deployed more quickly.

Manufacturing and Contributions

So we know that it’s an open source platform and anybody can manufacture this hardware by meeting OPC standards and distribute it across. In order to build up the concept of OCP hardware by Facebook, it was necessary for them to build hardware and prove to the world. Hence they have tied up with different hardware manufacturers to manufacture it. 

OCP hardware comes from a variety of Taiwanese hardware vendors, according to Cliff Grossner, the executive director of research and analysis at IHS Markit's cloud and data center research practice. These include vendors such as Wiwynn, Quanta Computer, Edgecore Networks, and Delta. But there are non-Taiwan vendors, such as Ericsson, that are also making gear.

That’s not to say there is no interest from the big players. When the time has run and slowly the major hardware vendors also started to contribute to this area(HPE, NOKIA,DELL, LENOVO, etc.). “For the most part, on the server side, we don’t see much interest,” said Grossner. “But on the networking side, Dell and HPE have OCP certified lines, we’re already seeing Cisco and others starting to use their silicon, and Mellanox has a good line of open network switches.”. 

Sourcing Open Compute Project hardware is getting easier. The project website features a marketplace through which you can research equipment specifications and contact Open Compute Project vendors. More of this details can be found in OPC website and here too.

OPC certifications & standards

The early focus on OCP servers and power supplies has grown to encompass into multiple devices and facilities such as,

·      Compute

·      Storage

·      Networking

·      Telco

·      Data Center Facility

·      HPC

·      Rack & Power

Basically, OCP has mainly three types of standard certifications that are, OCP Accepted, OCP Inspired, OCP Ready. 

Products that carry the OCP Accepted™ recognition comply 100% with an OCP accepted specification and the design files are open sourced and available.

Products that carry the OCP Inspired™ recognition comply 100% with an OCP accepted specification and are available from a Gold, Silver or Platinum member of OCP.

OCP Accepted™ or OCP Inspired™ products have been demonstrated and meet 3 or more of the OCP tenets-efficiency, openness, impact & scale. They can only be sold with warranty and support.

OCP Ready™ is not a certification for any of the products. This is the certification for the data center facilities. The OCP Facility Recognition program is intended to brand your facilities to be OCP Ready™. There are certain guidelines created by  OCP Data Center Facility Project Team and serve as a reference for data center operators and tenants who want to understand the fundamental facility requirements to deploy this gear into their IT space. Facilities that meet these guidelines and approved by the OCP DC Facilities Project receive the certification as an OCP Ready™ facility.

There are over 100 products referenced that have achieved either OCP Inspired or OCP Accepted recognition. The OCP Inspired label can only be used on products that comply fully with an existing, accepted OCP specification and that are made by an OCP Silver, Gold or Platinum member. OCP Accepted products can be made by anyone, but they too must comply fully with an existing, accepted OCP specification and open-source design files must be made available for them.

Advantages of the OCP

Open Compute offers a wealth of advantages for data center managers, administrators and consultants. An aim of Open Compute is to make data hardware consistent in effort to boost efficiency and minimize humor error. 

Greater control

The simplest way that we can say the same is there is ‘No Vendor Specific Locks’. Since vendors produce the same hardware from the same specs, members don’t have to stew over product discontinuation or collecting the correct parts. For example, let say that Facebook creates a compute server with higher performance and computing capability. Some other day other tech giants like Google may come and adopt this server technology and they will try to add more of the features/eliminate faults that they can. This is really an advantage in between all operations and growth. This is only because OCP is an open-source and there are no vendor specific locks on all devices or facilities included on this.

Savings

Another benefit of Open Compute project is savings. The project is made to slash operating expenses in comparison to conventional hardware by eliminating gratuitous equipment from data centers. Facebook’s Prineville Data Center, for instance, uses 38% less energy at 24% the cost in comparison to other data centers.

Scalability

Furthermore, Open Compute enables scalability. An individual design can be tailored to support various applications, such as high-performance computing and storage server applications. Upgrades are cheaper and simpler since all the components do not have to be replaced. Moreover, OCP data center components can be re-deployed for various applications, which minimizes expenses anchored to application-specific systems.

Modular design offers flexibility

The idea behind the highly modular design of an open compute server is that components can be swapped out as computing needs change. Besides reduction in cost (due to less customization), and flexibility in design, modularity offers other benefits such as augmentation (adding new solution by merely plugging in a new module), and exclusion (one of the goals is to move away from the wastage normally associated with a server upgrade). It says that due to this designs, it takes a maximum of 160 seconds to change any server component and 80% of other components in just 60 secs and all this without tools.

No tools required for regular maintenance

Open Compute Servers are designed with easy maintenance in mind. Not only is front serviceability providing easy maintenance for the engineer – there are also no tools required for regular maintenance work. Components can therefore easily and quickly be replaced, saving precious engineering resources.

Less components, less weight

The board on a server is a barebone design, where many common features are removed (as they are not considered needed). Examples of such are monitor ports and multiple expansion slots. Clever workaround designs are made to leave out components without losing functionality or adding costs. A power supply? There is no power supply on an OCP server, as it gets powered through a direct interface (from a power bar on the rack). A front bezel on a server usually fits no functional purpose, it actually limits the airflow through the server – therefore you won’t find it on an open compute server design.

All in all, the open compute’s vanity free design takes away on average 6lbs of unnecessary hardware on a server.

A final asset of OCP touched on earlier concerns the nature of open source projects. Since open-source projects are a group effort, the community leverages the talent of multiple organizations instead of one. Solutions to problems can be resolved quicker whenever a body of experts are focusing on the same issue.

Barriers of adopting OCP

In its 2018 spending study, IHS Markit identified the three main barriers to the adoption of OCP hardware as being concerns about security, sourcing, and integration.

One of the risks of giving everybody the specification to make OCP hardware is that anybody can make it: Somebody could tamper with it before delivery, and nobody would be any the wiser. In other words, supply chain security becomes a problem.

Security

At the OCP Summit held in San Jose in March 2018, OCP leaders said they were addressing supply chain security with the creation of a new Security Project focusing on the creation of a standard hardware interface and protocols for ensuring boot code integrity.

Microsoft has already contributed its Project Cerberus, a hardware root of trust for firmware on the motherboard designed to comply with NIST 800-192, Platform Firmware Resiliency Guidelines. 

Building on this base, they also plan to develop security firmware APIs, open-source firmware for dedicated security hardware, secure firmware provisioning methodologies, and tools to secure and verify all mutable storage, including flash for BIOS, microcontrollers and complex programmable logic devices (CPLDs). In this way, enterprises taking delivery of OCP hardware can be sure it’s only running the firmware they expect it to be running.

The project leads aren’t just concerned about new hardware: They’re also thinking of the second-hand gear. To secure the resale market, they will look into providing tools for recovering hardware from a compromised or untrusted state, and for tracking and changing its ownership.

Hardware-software integration

Integrating hardware and software is getting easier – particularly at the operating system level since Microsoft joined the OCP board and contributed designs for the racks and servers it is now using to deliver Azure services to its customers.

There is still work to be done at other levels, including the very lowest, the firmware that enables OCP servers to boot up.

That’s where another new OCP initiative comes in: the Open System Firmware Project. It’s working on open-sourcing the code that initializes server chipsets so that it can be used on a variety of platforms and processor types. Building on projects like UEFI and Linux Boot, it aims to provide support for all cloud operating systems and processor architectures found in the data center, including GPUs, FPGAs and other hardware optimized for applications such as machine learning.

With Open Compute hardware increasingly finding a role in network virtualization, there are also moves afoot to integrate open software and hardware here, too. Traditional networking equipment vendors like Cisco Systems or Juniper Networks tightly link the two, delivering proprietary software tailored to proprietary hardware.

OCP is working with the Linux Foundation to integrate its hardware with that organization’s Open Platform for NFV (OPNFV) software, and the two recently renewed their commitment to joint testing of hardware and software products meeting their respective specifications.

Summary

A lot of good has come out of OCP, but there is more work to do to close the gaps. And what is right for Facebook and other hyper-scalers is not necessarily right for a smaller scale datacenter. There are other avenues to consider for those looking to move to open, cost-optimized design principles that may address some of the shortcomings seen with OCP today. Some of the specialty vendors who sell OCP hardware have broader portfolios that may be a better fit for organizations outside the hyper-scale category. Also, large global vendors like Dell, Hewlett Packard Enterprise and Lenovo have business lines that provide hyper-scale-inspired designs that come with the service and support needed for many mainstream enterprises IT organizations and smaller service providers.

Knowledge Credits: www.opencompute.org , www.forbes.com

Have a comment or points to be reviewed? Let us grow together. Feel free to comment.