Open Compute Project

Announcing the OCP Compliance and Interoperability Project

Wednesday, January 09, 2013 · Posted by at 8:11 AM

As more and more organizations evaluate and adopt technologies based on Open Compute Project designs, it becomes increasingly important to ensure that those solutions are reliable and operate in an expected manner. To help meet that need, we're launching a new Compliance and Interoperability (C&I) project.

The C&I project is the seventh official OCP working group, and it will focus on establishing a transparent and scalable framework to ensure that solutions meet OCP standards for compliance and interoperability and are branded accordingly. This project will allow providers to leverage a common set of tools and processes for delivering solutions that meet OCP-defined expectations. The more thoughtful and robust the approach, the better for the OCP community.

And this is where you come in.

We invite you to join us at the fourth Open Compute Summit on Jan. 16-17 in Santa Clara, where we will host a Compliance and Interoperability technical track focused on discussing he most effective ways to certify OCP solutions. There is a vast amount of collective certification expertise within the OCP community, and the Summit offers a wonderful opportunity to begin harnessing that expertise for the greater good. We will also be looking for leaders to help us drive forward projects that will emerge out of the C&I project.

We look forward to seeing you at the OCP Summit. In the meantime, feel free to share any questions or comments by contacting us at opencompute-ci@lists.opencompute.org.

Eric is Vice President of Data Center Services at Fidelity Investments.

Continue reading this post

OCP Summit IV: Breaking Up the Monolith

Wednesday, January 16, 2013 · Posted by at 10:00 AM

It's hard to believe that the Open Compute Project was founded just a little over 18 months ago. At our first Summit, in June 2011, we had 200 participants, one official member (Facebook), and exactly three technology contributions (all from Facebook). Today, as nearly 2,000 people converge on Santa Clara to kick off the fourth OCP Summit, we have more than 50 official members and dozens of contributions from a wide variety of technology suppliers and consumers.

In October 2011 we announced the formation of a non-profit foundation to steer the project, and that organization has continued to evolve. In the last six months we've hired our first full-time employee, COO Cole Crawford, and added a new Compliance & Interoperability project. We've also seen our first international expansion, with the launch of an OCP Asia-Pacific chapter in Tokyo, and some encouraging engagement with the academic and research communities -- including a Purdue University competition to design a biodegradable server chassis.

We've also seen some tremendous engagement in the broader OCP community in the last six months. Some examples include:

 

 

    • New members: More than a dozen new organizations have joined, including storage companies like EMC, Fusion-io, Hitachi, and Sandisk; microprocessor organizations like Applied Micro, ARM, Calxeda, and Tilera; and communications companies like NTT Data and Orange.




    • Adoption of open source hardware: More companies are being inspired to think differently about how they build out their infrastructures. Examples include Riot Games, which has decided to purchase systems based on OCP designs and built by Hyve Solutions, and Rackspace, which has decided to design and build its own infrastructure using OCP designs as a starting point.




    • Momentum around Open Vault and Open Rack: We launched these two projects at the third OCP Summit in May 2012, and we're already seeing companies iterating on them in interesting ways and developing creative new technologies to complement them. Examples include new Open Rack designs from Avnet, Delta, Emerson, and Sanmina (many of which are being demonstrated at today's Summit) and new server designs for the SAS expander slot on the Open Vault spec from Calxeda and Intel.




    • Products being delivered: Also at the May 2012 Summit, AMD and Intel announced that they'd been collaborating with Fidelity and Goldman Sachs to develop new boards to suit those companies' specific workloads -- codenamed Roadrunner and Decathlete, respectively. Just seven months later, both companies are here today demonstrating finished products.




    • New technology contributions: We're starting to see significant momentum building behind technology contributions to the project. Some of the new contributions include designs for Fusion-io's new 3.2TB ioScale card; the designs Rackspace has created for its new infrastructure; and designs for a new storage box that Hyve Solutions built for Open Rack. Facebook is also making a few new contributions, including modifications to Open Rack and Open Vault for use in cold-storage environments and designs for a new, all-flash database server (codenamed Dragonstone) and the latest version of its web server (codenamed Winterfell).

 

But most exciting of all are a series of new developments that will enable us to take some big steps forward toward better utilization of these technologies. One of the challenges we face as an industry is that much of the hardware we build and consume is highly monolithic -- our processors are inextricably linked to our motherboards, which are in turn linked to specific networking technology, and so on. This leads to poorly configured systems that can't keep up with rapidly evolving software and waste lots of energy and material.

To fix this, we need to break up some of these monolithic designs -- to disaggregate some of the components of these technologies from each other so we can build systems that truly fit the workloads they run and whose components can be replaced or updated independently of each other. Several members of the Open Compute Project have come together today to take the first steps toward this kind of disaggregation:

 

 

    • Silicon photonics: Intel is contributing designs for its forthcoming silicon photonics technology, which will enable 100 Gbps interconnects -- enough bandwidth to serve multiple processor generations. This technology also has such low latency that we can take components that previously needed to be bound to the same motherboard and begin to spread them out within a rack.




    • "Group Hug" board: Facebook is contributing a new common slot architecture specification for motherboards. This specification -- which we've nicknamed "Group Hug" -- can be used to produce boards that are completely vendor-neutral and will last through multiple processor generations. The specification uses a simple PCIe x8 connector to link the SOCs to the board.




    • New SOCs: AMD, Applied Micro, Calxeda, and Intel have all announced support for the Group Hug board, and Applied Micro and Intel have already built mechanical demos of their new designs.

 

summitIVblog photo2

Taken together, these announcements will enable data center operators to build systems that better fit the workloads they need to run and to upgrade through multiple generations of processors without having to replace the motherboards or the in-rack networking. This should in turn enable real gains in utilization and unlock even more efficiencies in data center construction and operations.

There's a lot of work left to do here, but it's never been more important for us to take these steps. As our lives become more connected, and as more devices and applications generate more data, we will face compute and storage challenges that existing technologies cannot handle efficiently.

But we have one big advantage, as we face these challenges: We are doing this work together, in the open, and everyone has a chance to contribute -- to help ensure that all the technologies we develop and consume are as scalable as possible, as efficient as possible, and as innovative as possible.

Continue reading this post

Open Compute Summit IV Agenda

Wednesday, January 09, 2013 · Posted by at 11:11 AM

We've finalized the agenda for the fourth Open Compute Summit, starting next Wednesday, January 16, at the Santa Clara Convention Center.

You can see the list of keynote speakers and read the rundown of executive sessions and technical tracks on the summit's agenda page.

See you next week in Santa Clara, California!

Continue reading this post

Crowd-sourced solutions from GrabCAD

Thursday, January 24, 2013 · Posted by at 10:53 AM

Sara Sigel from GrabCAD was walking around consulting with teams when the JBOD team said they were having issues finding CAD models of a mini-SAS connector. She knew this would be a perfect task for the GrabCAD community. This is her story:

I posted a request with the drawings they had and asked for help. Within a few hours one of our community members let me know that 3M had all the CAD files they needed, deeply buried on their site!

The models were not in our library directly, which isn't a problem because it's the power of our community that gets the job done. Our resources are growing each day and the goal is to make engineers' lives better with stories just like this.

Sara Sigel is the community manager at GrabCAD.

Continue reading this post

Open Compute Summit IV: Hardware Hackathon

Thursday, January 24, 2013 · Posted by at 11:00 AM

Last week at the fourth Open Compute Summit in Santa Clara, we held the first-ever OCP hardware hackathon. In just nine hours, a group of about 70 hackers generated seven really interesting hardware designs — all of which could now be turned into physical products that are ready to plug in and power up.

Like in a software hackathon, the goal in a hardware hackathon is to take an idea from concept to prototype in as little time as possible. The big differences in a hardware hack are that the "prototype" is a set of manufacturing files (bill of materials, Gerbers, and solid models) and that projects are often judged not just on their innovation and utility, but also on how easily they could eventually be manufactured. At stake in this hack: The OCP Foundation would help the winning team patent their design or take it to prototype.

We began the process several weeks in advance, soliciting project ideas from the OCP community and zeroing in on six themes: storage, sensors & displays, communication, power, racks & mechanical, and robotics. My company, Upverter, provided design tools for the teams to use, and GrabCAD got involved to fill out the mechanical side of the hack.

And then came the main event! At 1:00 pm on the first day of the Summit the 70 hackers got together in one of the ballrooms. We walked through what tools were available and what type of files could be submitted at the end of the hack. We then started a timer and the hackers began to find and form teams.

The energy that poured into ideation and team formation was pretty awesome to watch. We had whiteboards set up all over the room, with little semicircles around each and hackers debating the problems they were excited about. After about 30 minutes teams were formed, ideas were picked, and the actual hacking began.

Here are the projects the different teams tackled:

19" to Open Rack conversion

The challenge: Build mechanical and power interoperability between legacy compute hardware and the Open Rack standard. The hack was predominantly mechanical and resulted in a three-piece rack design that could be configured to accept either type of hardware. The team was working on the mechanical designs at the end of the hack.

Bluetooth adaptor

The challenge: Get rid of the USB and serial cable mess that’s required to hook a laptop up to the debug port of a server. The hack designed a small circuit that allowed the technician to connect to a server via Bluetooth, instead of via a serial cable. The team had a working prototype and design files in Upverter at the end of the hack.

Debug port aggregation

The challenge: Collect all the debug port information from all the servers in a rack and connect it to the Internet. The hack was designed mechanically and electrically and was mostly complete by the end of the hack.

The "FlexiStrip"

The challenge: Line a smattering of server racks from floor to ceiling with an array of temperature sensors and then use that data to generate a thermal profile of the data center. The team focused on the sensor arrays themselves, but they also did some Arduino prototyping into reading sensors and wireless communication.

The "Happy Rack"

The challenge: Monitor the health and happiness of a server rack and display it prominently and physically on the top of the rack. The hack was done as a prototype cardboard box with an Arduino-driven arrow that moved between an unhappy face and a happy face depending on input. (This hack won our unofficial "Hacker Spirit Award.")

High-density Open Rack JBOD server

The challenge: Take a high-density array of cheap disks and design an enclosure for them that would fit into an OCP Open Rack. The team made it as far as mechanical and thermal models. They were beginning to work on the carrier/backplane circuit boards at the end of the hack.

Mesh-networked sensors

The challenge: Distribute a large number of low-power sensors around a data center and communicate between them efficiently. The hack was built on top of Arduino, ZigBee wireless, and a simple thermal sensor. The team focused on a prototype, finished it early, and spent the rest of the hack on documentation and moving the design into Upverter.

And the winner is…

At the end of the hackathon, all the groups did quick presentations on their hacks. We were all impressed by the quality, applicability, and completeness of all of the projects. To determine the winner, we looked at the five deliverables (schematic, layout, mechanical, prototype, documentation), the hack's applicability to data centers, and the hack's completeness and manufacturability.

The winning design: The debug port aggregation hack. This team produced a design that hit 4 of the 5 deliverables, was super applicable, and was just hours from completion.

Congratulations to all the teams involved for their great work, and see you at the next OCP hardware hack!

Zak Homuth is the co-founder and CEO of Upverter. Upverter builds the pickaxes of hardware engineering. Ideation tools, schematic capture, PCB layout, design hosting, parts library management, and affordable prototyping. All of which are collaborative, accessible, reusable, and in the cloud.

Continue reading this post