Open Compute Project

Anita Kou

Friday, May 29, 2015

 

Anita06JP

Ms Anita Kuo holds EMBA (in Global Business) & LL. M (in Privacy Law and ICT Standards) degrees from National Chengchi University, Postgraduate Diploma in Business Systems from Monash University, Bachelor of Administration (in International Trade) from National Taiwan University. Anita joined ITRI in 2006. With substantial experiences in both industry and government ICT projects for 20 years, as well as in multi-national organizations (APEC TEL, Asia PKI Consortium, Pan-Asia e-Commerce Alliance, etc.) & global ICT ecosystem linkage, Anita represents OCP Taiwan as a principal liaison between the industry and government, and interacts with Taiwan cloud ecosystem for CCMA (Cloud Computing Research Center for Mobile Applications), ITRI and CCAT (Cloud Computing Association in Taiwan).

ITRI's establishment of OCP Taiwan, non-profit status and proximity of key vendors provide unique capabilities for the OCP ecosystem. ITRI commits to support OCP in the fullest extent possible. Anita has been appointed by ITRI as the lead liaison role on promoting OCP to the industrial ecosystem.

Anita co-works with Chilung, Director of ITRI OCP Certification Testing Center, the OCP Incubation Committee Hardware Management Seat, just like partners complement to each other’s strength. Moreover, UNH-IOL tag-teams with ITRI. With such capacity, Anita and David each are not only willing to take the interim role, further leverage ITRI’s technology and industrial influence with the synergy of UNH-IOL’s domain influence in OCP ecosystem, but also to contribute the time and every effort as the next project lead of OCP Certification.

Continue reading this post

David Woolf

Friday, May 29, 2015

 

 

David Woolf

 

David Woolf currently leads Data Center initiatives at the University of New Hampshire InterOperability Laboratory (UNH-IOL).

While at UNH-IOL, David has led development of dozens of industry-reviewed test programs and procedures as part of the team that has grown the UNH-IOL into a world-class center for interoperability and conformance testing.

David has been an active participant in a number industry forums and committee’s addressing conformance and interoperability, including the SAS Plugfest Committee, SATA-IO Logo Workgroup, co-chair of the MIPI Alliance Testing Workgroup, and coordinating the NVMe Integrators List and plugfests. His dedication to neutrality, and open specifications standards has served him well in these fields.

In line with this David has been instrumental in architecting and executing interoperability events around the world, including industry first plugfests in SAS, iSCSI, MIPI, and NVMe.

Continue reading this post

Matt Peterson

Friday, May 29, 2015

 

20130530 Matt Straight

 

Matt Peterson works within the office of the CTO at Cumulus Networks. At Cumulus, he built the initial customer experience team, with more recent responsibility around technical evangelism and customer product direction. Matt has held enable access on 2 to 5 digit ASN organizations - as well as co-founding the first non-profit IX within San Francisco, known as SFMIX. His work has been presented at numerous industry events, including APRICOT, BSDcon, Defcon, & CCC Camp. Within the OCP context - he's passionate about enabling a DevOps mindset within Open Hardware. To this end, his efforts have been spread around ONIE adoption, pluggable optics testing, and hardware management.

Continue reading this post

OCP C&I Project Special Election

Wednesday, May 06, 2015 · Posted by at 2:18 AM

The OCP Community is an ever evolving, innovating and thriving community. In March we announced that we would be extending the terms of our volunteer leadership and holding no elections until 2016. However, we now have special circumstances requiring a special election.

Special Circumstances

If an OCP project lead resigns from their position prior to the end of the term or if they can no longer fulfill their duties then the OCP project community will vote on the interim project lead.

Special Elections

YF Juan has announced his resignation as the C&I project lead.  We appreciate all his hard work and dedication to not only the OCP C&I project, but to the Foundation as well. His last day as project lead will be this Friday, 8 May 2015.  As such today, 6 May 2015  through 16 May 2015 we will be taking nominations for this position.

Time Line for the OCP C&I Project Lead Special Election

6-16 May 2015

Nomination Phase

16-19 May 2015

Validate Eligibility

20 May 2015

Contact Nominees

21-31 May 2015

Testimonials

1-13 June 2015

Voting Phase

16 June 2015

Results Announced

Roles and Responsibilities of the an OCP Project Lead

Leadership of an OCP project is, by far, one of the most important volunteer roles within the OCP ecosystem.

The role of an OCP Project Lead includes (but not limited to):

  • facilitate discussions, both online and in-person, that empower and encourage project participants to communicate, collaborate, and contribute to OCP in an open and transparent manner;
  • support the direction of the Foundation by effectively communicating those goals to their projects through the OCP mailing lists, monthly calls, OCP events and various other tools;
  • drive communication between their projects and the Foundation;
  • work closely with the Foundation to grow the reach of their respective projects and that of the OCP ecosystem;
  • focus on the needs of the project participants, the projects goals, and the OCP Foundation’s goals and philosophies.  

In addition the Project Leads must be able to infuse their project with the following traits:

  • strong community leadership
  • world wide collaboration
  • inclusive vs exclusive approach
  • technical excellence
  • focus on innovation

How does nominations work?

Any member can nominate themselves or someone else on the project for inclusion on the ballot. Please fill out the webform on or before 5pm Central time on 16 May 2015.

Who is eligible for Nomination?

Any individual member or anyone from a member organization whose membership is in good standing.  All nominees memberships will be validated prior to 19 May 2015. You must create an individual profile on the website prior to 15 May 2015 to be eligible. If you belong to a member organization please use your organization email to create this profile.  Your organizational membership takes precedence over the individual membership; however, we pull the list of names from this database.  Instructions for creating this profile can be found at:http://www.opencompute.org/community/get-involved/tiered-membership/individual-ocp-community-membership/

Terms

The interim project lead will hold this seat until the regular scheduled project lead election in August 2016.

Who is eligible to vote?

The OCP Board members as well as any individual member or anyone from a member organization whose membership is in good standing.  You must create an individual profile on the website prior to 25 May 2015 to be eligible to vote. If you belong to a member organization please use your organization email to create this profile.  Your organizational membership takes precedence over the individual membership; however, we pull the list of names from this database.  Instructions for creating this profile can be found at:http://www.opencompute.org/community/get-involved/tiered-membership/individual-ocp-community-membership/

Restrictions and Limitations:

  • To be eligible for nomination you must create your profile on the website on or before 15 May 2015.
  • All nominations must be added to the webform no later than 5pm central on 15 May.
  • You must be an individual member (by creating your profile you are signing the individual agreement) or an employee of a member organization in good standing in order to be nominated. Nominee membership will be validated by 18 May 2015
  • Individual, Community and Silver level members are eligible to hold no more than 1 leadership position at any one time
  • Gold level members are eligible to hold no more than 2 leadership positions at any one time
  • Platinum level members are eligible to hold no more 3 leadership positions at any one time
  • No person shall hold 2 leadership positions at any one time
  • Leadership positions are defined as OCP Board Member, OCP Incubation Committee (IC) member (excludes chair and vice-chair), OCP Project Leads and OCP Approved Regional Community Leaders.
  • You must be an individual member (by creating your profile your are signing the individual agreement) or an employee of a member organization in good standing on or before


Many thanks to YF Juan for all his hard work on the OCP C&I project and we look forward to seeing the community nominees.

If you have any questions please email Amber Graner.






Continue reading this post

OCP U.S. Summit 2015 Video and Presentation Links

Friday, May 08, 2015 · Posted by at 9:51 AM

OCP U.S. Summit 2015 was held on 10-11 March 2015 at the San Jose Convention Center in San Jose California, This year's summit had 49 sponsors and over 2500 registered attendees which represented over 800 companies from 40 countries. If you were unable to attend OCP U.S. Summit 2015 below are links to the main stage talks, slides as well as pictures from the event.

Main Stage Talks

Day 1 - 10 March 2015

  • Keynote: Welcome - Open Compute Updates - Frank Frankovsky, Chairman & President - Video, Slides

  • Keynote: Hewlett-Packard - Alain Andreoli, Senior VP, John Gromala, Senior Director, Antonio Neri Senior VP - Video, Slides

  • Keynote: Keynote: Facebook News - Jay Parikh, VP of Engineering, Facebook - Video, Slides

  • Keynote: Mark Shuttleworth, Founder of Ubuntu and Canonical, Video

  • Keynote: Financial Services Adoption Panel - Peter Krey, President & Founder, Krey & Associates, Inc. -Video, Slides

  • Sponsored Theatre - Expo Hall Floor: Intel - Video

  • Executive Track: Networking Project Update - Omar Baldonado, Project Lead - Video, Slides

  • Executive Track: Data Center Project Update - Jason Schafer, Regional Director of Operations, Maicom Facility Services - Video, Slides

  • Executive Track: Renewable Energy Panel - Bill Weihl, Moderator - Video, Slides

  • Executive Track: Goldman Sachs Adoption Story - Don Duet, Managing Director - Video, Slides

  • Executive Track: OCP 101 - Amber Graner, OCP Community Manager - Video, Slides

Day 2 - 11 March 2015 

 

  • Keynote: Microsoft / OCP New Announcements - Kushagra Vaid, GM, Cloud & Enterprise Group - Video, Slides

  • Keynote: Broadcom Fireside Chat - Najam Ahmad, VP of Network Engineering, FB - Video, Slides

  • Keynote: Rackspace - Aaron Sullivan, Distinguished Engineer, Infrastructure Strategy - Video, Slides

  • Keynote: Sam Ramji, CEO, Cloud Foundry Foundation - Video, Slides

  • Executive Track: On Premise vs. Cloud - Amir Michael - Video, Slides

Pictures

 

Images from the event can be found on the OCP Flickr page. https://www.flickr.com/photos/106659057@N08/sets/72157651601950765/

 

Upcoming Summit

The next summit will be OCP U.S. Summit 2016 which will be held on March 9-10 2016 at the San Jose Convention Center in San Jose, California. We will update the OCP Summit page as soon as updates are available.

 

Continue reading this post

 

We will hold an OCP Telco Working Group Engineering Workshop on 8 June from 1-5pm at McCormick Place in Chicago, Illinois. This workshop will take place the day before Light Reading’s Big Telecom Event. The OCP Engineering Workshop is free and open to the public; however, registration is required. To register for the OCP Telco Working Group Engineering workshop click here.

What is an OCP Engineering Workshop?

The OCP Engineering Workshops are working sessions that are meant to be technical in nature and are designed to allow the OCP Project Communities to come together to work on the goals and set the direction of each project as it supports the mission and vision of the OCP Foundation.  These Engineering Workshops are designed to encourage, empower, and inspire the OCP community to communicate, collaborate and contribute! These events are free to the public, but do require registration.

The goals of these workshops include:

  • Increase community participation and stickiness.
  • Increase the OCP technology portfolio; depth and breadth of specifications and charters.
  • Drive the product-ionization of OCP’s specifications.  
  • Solidify specification versioning pipeline, release process, naming conventions.
  • Take into account/understand the suppliers needs and OCP partners needs
  • Implementation of OCP’s legal framework (CLA and OCPHLs).
  • Building out/addressing technologies for new sectors.

 

OCP Teclo Working Group Engineering Workshop

OCP Telco Working Group Engineering Workshop will be held on 8 June 2015  from 1 to 5pm in conjunction with Light Reading’s  Big Telecom Event summits and workshops which will take place at McCormick Place in Chicago, Illinois.

Agenda*:

1 - 1:15pm  - Welcome (Corey Bell, OCP CEO)

1:15 - 2:00pm - OCP 101 - What is OCP?  Why is it important?  How to contribute and consume OCP and more. (Amber Graner, OCP Community Manager)

2-3pm - What are the goals of the Telco Working Group? How you can help?  How to get involved. (Bill Carter, Intel)

3-4pm - Impact of Open Hardware (TBD)

4-4:30pm  - Success Stories (TBD)

4:30-5pm - General Q&A, Next steps, Wrap up (Bill Carter, Corey Bell, Amber Graner)

*Agenda is subject to change

Registration for the OCP Telco Working Group Engineering Workshop

Register today!  The OCP Telco Working Group Engineering Workshop is free and open to the public, but registration is required.  Click here to register for the OCP Engineering Workshop.

Venue

MCCORMICK PLACE

2301 South Doctor Martin Luther King Junior Drive, Chicago, IL 60616

More information about the venue, hotel, and travel arrangements can be found on Light Reading’s Big Telecom Event webpage.

Light Reading’s Big Telecom Event

Light Reading's Big Telecom Event (BTE) is the premier North American event in the telecom space featuring more than 80 top-notch service provider speakers from the world's leading telcos. BTE 2015 tackles the big issues facing the telecom industry and how service providers can make money from them.

 Special offer for OCP Workshop attendees: REGISTER TODAY for complimentary or discounted VIP admission. BTE provides complimentary admission to employees of service providers, operators, financial and educational institutions, utilities and government agencies. All others will receive a 30% discount. Enter promo code PTNR_OPENCOMPUTE.

VIP Admission Includes:

·         VIP Onsite Check-In

·         VIP Welcome Gift

·         The Heavy Reading Exclusive BTE Research Package: A 5,000 word strategic analysis of the state of the art in monetizing next-generation, high-capacity networks, produced exclusively for BTE VIP attendees.

Register Now to receive your complimentary pass and VIP admission. Visit Light Reading’s Big Telecom Event and Enter promo code: PTNR_OPENCOMPUTE

More information about BTE can be found here.

 

Continue reading this post

The 2015 Open Compute Project (OCP) Hackathon took place from noon on Tuesday, March 10, until noon on Wednesday, March 11. Each team had 24 hours to develop, implement and demonstrate a “hack.” The hacks use only opensource hardware and software technologies. The winning hack must be useful, novel and non-obvious.

In 2014, the winning team consisted of Dimitar Boyn from I/O Switch Technologies, Inc., Jon Ehlen from Facebook, Inc., Ron Herardian, a self-described computer hobbyist and hacker, Rob Markovic, an independent consultant, Peter Mooshammer from Opscale, LLC, and Andreas Olofsson from Adapteva, Inc. The winning hack, Adaptive Storage, was a radical new storage architecture.

At the 2015 Hackathon, the team was re-formed. The new team, dubbed “Hardcore Hackers,” consisted of Dimitar Boyn, Ron Herardian, Rob Markovic, Peter Mooshammer and Austin McNiff, a freshman student at San Jose State University. Most of the team members met for the first time at the 2014 Hackathon and none of the team’s members were previously acquainted with Austin McNiff.

The team’s 2015 hack was to implement n-way redundancy for the Zettabyte File System (ZFS) using OpenZFS and to scale it horizontally using flash memory switched fabrics. If it worked, the hack could revolutionize ZFS. Instead of relying on traditional two-way clusters to provide high availability, ZFS could provide continuous availability using commodity hardware.

ZFS is an opensource file system available in FreeBSD and, through the OpenZFS project, on Linux and Mac OS X, as well as on IllumOS, which is a continuation of OpenSolaris. The team used CentOS, which is an opensource distribution of Linux based on Red Hat Linux. Red Hat is widely used in data centers and the company plays a major role in many open source projects.

The 2015 OCP Hackathon became the first demonstration of a non-proprietary ZFS storage system with a single file system spanning more than two servers. It was also the first ever public demonstration of a native Peripheral Component Interconnect Express (PCIe) flash memory switched fabric.

 Flash Memory Switched Fabrics

Dimitar Boyn brought several flash memory switched PCIe fabric cards and dozens of I/O Switch Raijin New Generation Form Factor (NGFF) solid state drives (SSDs) to the event. I/O Switch is a corporate member of OCP and is in the process of opensourcing its Raijin SSD card design through the OCP Storage project. The team quickly assembled a lab environment on site at the event using commodity x86_64 servers and networking equipment.

The SSDs, manufactured by Memoright International, Inc., are designed for standard Mini PCIe version 2 (M.2) connectors used in notebook computers and mobile devices. However, the SSDs contain enterprise-grade Multi-Layer Cell (MLC) NAND (“Not AND”) flash memory and have advanced wear leveling and error correcting firmware. As a result, standard M.2 SSDs can be used in servers. Using standard parts based on open standards reduces cost.

The open standards based I/O Switch Thunderstorm flash memory switched fabric PCIe card has four M.2 connectors and can use either Advanced Host Controller Interface (AHCI) or Non Volatile Memory Express (NVMe) M.2 SSDs. In systems with an AHCI compatible Basic Input Output System (BIOS), which includes virtually all computers manufactured in the last decade, the SSDs operate as native, locally attached block devices. In the AHCI implementation, built-in driver support is available in every major operating system (OS), which reduces complexity.

2015 ocp hackathon n way openzfs redundancy 01 thunderstorm2 

Figure 1. I/O Switch Thunderstorm Flash Memory Switched Fabric PCIe Card

Native PCIe SSDs are much faster than Serial Advanced Technology Attachment (SATA) or Serial Attached Small Computer System Interface version 2 (SAS-2) SSDs because of flash memory de-indirection. The electrical and logical paths between application code and data are shorter and Input / Output (I/O) latency is lower because there is no disk controller bottleneck. Applications or software data protection technologies, such as Logical Volume Manager (LVM), software Redundant Array of Inexpensive Disks (RAID) or Just a Bunch of Disks (JBOD) can be used instead of a conventional Host Bus Controller (HBA).

Storage solutions can be optimized for I/O Operations per Second (IOPS), latency, throughput or capacity. Flash memory switched fabrics are optimized for throughput but provide competitive IOPS and latency. Some proprietary flash memory package approaches perform better in terms of IOPS, have higher storage density per PCIe slot, or offer lower latency, for example, by placing flash memory on Dual In-Line Memory Modules (DIMMS). However, proprietary flash memory package approaches are more costly compared to standard M.2 SSDs.

The ability to add many SSDs to a server without using any hard disk drive (HDD) bays is especially useful for storage solutions based on ZFS. In ZFS, read I/O operations are served from the Adaptive Read Cache (ARC) which resides in Random Access Memory (RAM). A Level 2 Adaptive Read Cache (L2ARC) is normally placed on SSDs so that rotating disk drives are accessed only as a last resort. With sufficient RAM and flash memory, after a cache warm up period, cache hit rates become very high for most application workloads, which improves performance for read I/O operations. In the hack, each server had five 512GB SSDs for a total of 2.5TB of flash memory per server.

 2015 ocp hackathon n way openzfs redundancy 02 cards in server1 

Figure 2. I/O Adding Multiple PCIe SSDs to a Commodity Server

OpenZFS on Flash Memory Switched Fabrics

As with other journaled file systems, write I/O operations are logged in ZFS. Writes are acknowledged as soon as they are committed to the log, which is much faster than waiting for other file system operations to complete, especially if the log is on a SSD while rotating HDDs are used for storage capacity. In ZFS, the log is referred to as the ZFS Intent Log (ZIL).

 

2015 ocp hackathon n way openzfs redundancy 03 zfs model1 

Figure 3. ZFS Model

A single flash memory switched fabric PCIe card can deliver SSDs for OS boot, journaling (ZIL), read caching (L2ARC) and additional capacity. For example, both boot and ZIL can be mirrored on a single card or they can be mirrored (RAID1), striped (RAID0) or mirrored and striped (RAID10) across cards with two or more cards. Since the cards are modular and use standard parts, different sizes and types of SSDs can be combined as needed. SSDs can normally be upgraded or replaced in the field without removing the card from its slot.

Implementing n-way ZFS redundancy and horizontal scaling using ZFS, requires abstracting the underlying block devices away from the hardware before adding them to ZFS. Internet Small Computer System Interface (iSCSI) was used to abstract block devices used to construct ZFS virtual devices (“vdevs”) and Distributed Replicated Block Device (DRBD) was used to abstract the block device used for the ZFS file system’s log.

 

2015 ocp hackathon n way openzfs redundancy 04 zfs devices1 

Figure 4. ZFS with Virtualized Block Devices

Using DRBD for the ZFS Log Device

ZFS is not a clustered file system and is not designed to scale horizontally. Redundancy is normally provided by configuring two storage servers connected to the same SAS-2 drive enclosures in a traditional two node cluster. When one server fails, the other server takes over the storage devices and begins providing the same Network Attached Storage (NAS) shares or Storage Area Network (SAN) Logical Unit Numbers (LUNs).

To ensure that uncommitted writes are preserved during a server failure, the ZIL must be attached to the SAS-2 bus, which is much slower than native PCIe storage devices. Redundancy in ZFS can be provided within vdevs or using RAID, but there can only be two storage heads. If both servers fail, the storage resources are unavailable until at least one server is repaired.

Creating horizontal redundancy for three or more servers requires replicating the ZIL across machines. The solution was to define the ZIL as a DRBD.

 

2015 ocp hackathon n way openzfs redundancy 05 drbd cli1 

Figure 5. Distributed Replicated Block Device Consisting of Native PCIe SSDs

DRBD works on top of block devices, such as hard disk partitions or LVM volumes. It synchronously mirrors each block of data that is written to the device to one or more peers over a network using Transmission Control Protocol / Internet Protocol (TCP/IP). The file system on the active server is notified that the write has completed only when the block is successfully written to all disks in the cluster. Using ZFS, data can be accessed using a ZFS file system only on the active node.

 

2015 ocp hackathon n way openzfs redundancy 06 drbd1 

Figure 6. Synchronous Mirroring of ZFS ZIL using DRBD

For a non-primary DRBD node to use the DRBD ZIL in ZFS, all of the DRBD peers have to be configured as primary nodes. DRBD’s dual-primary mode allows concurrent access to the device, thus a clustered file system, such as Global File System (GFS), that utilizes a distributed lock manager is required. To eliminate the need for a distributed lock manager, the DRBD device was not added to a ZFS pool (“zpool”) on the non-primary nodes.

 2015 ocp hackathon n way openzfs redundancy 07 robi and austen1 

Figure 7. Rob Markovic (pointing) and Austin McNiff (foreground)

Once DRBD was working, the rest of the data in the file system had to be mirrored across servers.

Using iSCSI to Create Mirrored ZFS Virtual Devices

Data was mirrored using iSCSI, which allows block device commands and data to be sent over a network using TCP/IP. Storage devices exported by one server can be attached by another server where they appear as local block devices. For the hack, all of the drives used for data were exported as iSCSI LUNs, including the local disks in the primary server, and then added to ZFS vdevs configured as mirrors where each mirror spans multiple nodes.

2015 ocp hackathon n way openzfs redundancy 08 iscsi cli1 

Figure 8. ZFS zpool with Mirror iSCSI LUN vdev and DRBD Log Device

The configuration ensures that both the data and the write log are mirrored across all of the servers. The file system can be read by the primary server, which shares its resources over the network using Network File System (NFS).

In a real system, the IP address of the NFS server could be virtualized, for example, as a floating IP address, so that after a failure of the primary server, the non-primary servers would continue operating when a new primary server takes over.

Each non-primary server mounts the NFS shares on the primary server. Reads by the non-primary servers take advantage of Remote Direct Memory Access (RDMA). Once the read cache on the primary server warms up, reads are served directly from ARC which is accessed by the network interface card on the primary server through RDMA. Write I/O operations also use RDMA via iSCSI Extensions for RDMA (iSER). In a real system, a 10Gbps Ethernet or InfiniBand network would be used to reduce the latency of I/O operations.

Promoting a Non-primary Server

If the primary server is down, a non-primary server can log in to the iSCSI targets. The iSCSI block devices are only mounted when the zpool is imported. Since DRBD preserves the logical names of devices across nodes, “zpool import –f” (force import) can be used to make a non-primary server the new primary server (the OS instance currently controlling the file system). When the zpool is imported, uncommitted writes in the ZIL can be recovered.

In the Hackathon implementation, the L2ARC on the new primary server had to warm up because the read cache was not replicated. However, since the zpool metadata contains the information for both the ZIL and the L2ARC, it is also possible to replicate the L2ARC using DRBD and to provide a hot read cache failover capability.

In a real system, the steps involved in detecting the failure of the primary server, importing the zpool on the new primary server and sharing its file systems using NFS would be automated.

Under normal operating conditions, the Central Processing Units (CPUs) on the non-primary servers are largely idle. As a result, the hack is ideal for a cluster of application servers or for an object store. For the hack, an Apache web server was used as a proxy for an object store Application Programming Interface (API).

The key insight, provided by Dimitar Boyn, was that access to data in the primary server’s ARC via RDMA running over InfiniBand will be faster than access to local SSDs in the same machine, even native PCIe SSDs. Reads in the underlying iSCSI horizontal mirrored vdevs use iSER. The latency of InfiniBand and of Direct Memory Access (DMA) is much less than the latency of flash memory. Therefore, the hack is a read optimized solution where appropriate use cases will have a majority of reads. The only thing faster would be reads from local ARC.

Since the servers used in the hack were basically all-flash arrays, there was never any concern about the latency of accessing rotating disk drives. While the primary server can simply serve data from ARC and L2ARC after a cache warm up period, the non-primary servers can run applications. At the same time, each non-primary server has a complete copy of the data in standby mode.

The Moment of Truth

The hack took all night and into the next day. Most of the team rotated in and out, working on various problems late into the night. With the help of Ron Herardian and Rob Markovic, Dimitar Boyn (Founder and Chief Technology Officer of I/O Switch) pulled an all-nighter to make sure that the failover process worked as expected.

 2015 ocp hackathon n way openzfs redundancy 09 dimitar1

Figure 9. Dimitar Boyn after Working Through the Night

The team worked right up until the hack was presented. All of the team members participated in the presentation, except Austin McNiff, who was unavailable due to his class schedule. As a demonstration, a web browser was shown on a laptop accessing media files on an Apache web server running on a non-primary server while the media files were accessed using NFS.

 

 2015 ocp hackathon n way openzfs redundancy 10 solution before1  2015 ocp hackathon n way openzfs redundancy 11 solution after1

Figure 10. Before a Server Failure

Figure 11. After a First Server Failure

 In the hack, there was a single primary server and writes were mirrored from the primary server to the non-primary servers. If the master failed, any non primary server could become the new master. Topologies with more than one primary server are also possible. In a ring topology, each server is primary for one file system and non-primary for other file systems in the ring. Any non-primary server can take over any file system in the ring. Each server has an active zpool controlling a subset of SSDs in its mirrored vdevs and DRBD ZIL device and shares its file system using NFS.

 

2015 ocp hackathon n way openzfs redundancy 12 2d torus1 

Figure 12. Two Dimensional Torus

In a four server ring, the patterns formed by the replication paths of data mirrored across vdevs and ZIL devices combined with the patterns formed by alternate forward or backward failover sequences are described by a two dimensional torus.

The 2015 OCP Hackathon proved that open technologies and open collaboration greatly accelerate innovation. Prior to the 2015 OCP Hackathon, there was no non-proprietary way to scale OpenZFS beyond two servers. Horizontal scaling could only be accomplished by placing multiple copies of data on a series of separate storage systems, each of which had only single redundancy. Using opensource software and open standards based hardware, n-way ZFS redundancy and horizontal scaling was implemented by a small team of hackers in just 24 hours.

 

Continue reading this post

Integrated Device Technology, in collaboration with the Centre for Business Analytics and Performance at the Telfer School of Management, University of Ottawa will be hosting an OCP Day on June 18th, 2015. During its OCP Day events, the OCP Foundation strives to promote the adoption of OCP accepted specifications and designs, to inspire innovation in, near and around the data center ecosystem and beyond, and to encourage participation in the OCP community in an open and transparent manner.

The objective of this event is to discuss the formation of an OCP Canda chapter focused on High Performance Computing (HPC). The benefits and potential of HPC is recognized by many. At the moment however, this capability is limited to large computing centres. Imagine that HPC could be distributed to the edge of the network as opposed to the centre only. We could reduce latency, push analytics to the point of impact and thereby improve the use of data to solve transportation, natural resources, health care, emissions and other societal challenges. Once formed and approved the mission of the Canada based chapter will be to explore the ways and means of doing so collaborating with the global Open Compute community.

 

What is an OCP Day?

An OCP Day is a half-day event where the OCP Foundation and members of the community get together to introduce OPC to new organizations in a specific area of the world.

The event is free and open to the public, but registration is required.

 

OCP Day Details

Date: Thursday, 18 June 2015
Time: 1:00 - 5:00 PM
Location: University of Ottawa, Desmarais Building, 55 Laurier Avenue East, Ottawa, K1N 6N5
Registration: Is required, click here to register.

 

Agenda*

(*subject to change; all times local)

1:00-1:20 pm              Intro to Open Compute
Corey Bell, OCP CEO
     
1:20-1:40 pm              Intro to OCP HPC
Devashish Paul, OCP HPC Co Chair / Director of Strategic Marketing, IDT
     
1:40-2:00 pm   Opportunities for HPC in Solving Societal Problems
Dr. Greg Richards,MBA, Ph.D, FCMC / Director, MBA Program, Telfer School of Management
     
2:00-2:20 pm   How we Built a Multi-Vendor Next Generation Network in 120 Days
Ritch Dusome, President & CEO, CENGN
     
2:20-2:40 pm                  RapidIO.org Update on HPC and Supercomputing Initiatives
Rick O’Connor, Executive Director, RapidIO.org
     
2:40-3:00 pm   Coffee Break
     
3:00-3:30 pm   OCP Project Updates
Amber Graner, OCP Community Manager
     
3:30-4:00 pm   Creating a Low Latency HPC Cluster in Canada
Mohammad Akhter, Director of Architecture, IDT
     
4:00-5:00 pm  

Networking

Registration for this event is now open; register today! We hope to see you in Ottawa and look forward to your particpation in the OCP community.

Continue reading this post

The Testimonial Phase of the OCP C&I Project Lead Special Election is now open and the Foundation would like to thank all the nominees who have agreed to run for the OCP C&I Project Lead seat.

Leadership of an OCP project is, by far, one of the most important volunteer roles within the OCP ecosystem.

The role of an OCP Project Lead includes (but not limited to):

  • facilitate discussions, both online and in-person, that empower and encourage project participants to communicate, collaborate, and contribute to OCP in an open and transparent manner;
  • support the direction of the Foundation by effectively communicating those goals to their projects through the OCP mailing lists, monthly calls, OCP events and various other tools;
  • drive communication between their projects and the Foundation;
  • work closely with the Foundation to grow the reach of their respective projects and that of the OCP ecosystem;
  • focus on the needs of the project participants, the project goals, and the OCP Foundation’s goals and philosophies.  

 

In addition the Project Leads must be able to infuse their project with the following traits:

  • strong community leadership
  • world wide collaboration
  • inclusive vs exclusive approach
  • technical excellence
  • focus on innovation
  •  

As we enter the testimonial phase of this election, we are pleased to announce the following members of our community which have been nominated for the C&I Project Lead seat.

Starting today through 12 June you will be able to add a testimonial for your support of the various nominees. Voting for the C&I Project Lead seat will begin on 15 June and end on 26 June.  The results of the poll will be announced on 29 June.

To add a testimonial you will need to go to the page of the person you wish to endorse and add your support for that nominee to the comment box.

More information about our governance process can be found here.

Again, the Foundation would like to thank all of those who have been nominated.  None of our successes would be possible without the support and participation of you and your organizations.

Get Involved

The OCP Foundation has many ways for individuals and organizations to participate in OCP; online project meetings, OCP events, submit a spec, review a spec, read our blog, join our Facebook group, add us to your G+ circles, follow us on Twitter, join our mailings and much more. To find out more about getting involved in OCP, click here.

 

Continue reading this post