Archive

Archive for the ‘InfiniBand’ Category

Latest InfiniBand and RoCE Developments a Major Focus at the OpenFabrics Alliance Workshop 2018

May 31st, 2018

openfabriclogo

The annual OpenFabrics Alliance (OFA) Workshop is a premier means of fostering collaboration among those in the OpenFabrics community and advanced networking industry as a whole. Known for being the only event of its kind, the OFA Workshop allows attendees to discuss emerging fabric technologies, collaborate on future industry requirements, and address remaining challenges. The week-long event is made up of sessions covering a wide range of pressing topics, including talks related to InfiniBand and RDMA over Converged Ethernet (RoCE).

This year’s agenda featured sessions highlighting a variety of InfiniBand and RoCE updates and emerging applications. Below is a list of all OFA Workshop 2018 sessions covering RDMA technologies and the associated presentations.


RoCE Containers - Status update

Parav Pandit, Mellanox Technologies

Using RDMA in containerized environment in a secure manner is desired. RDMA over Converged Ethernet (RoCE) needs to operate and honor net namespace other than default init_net. This session focused on recent and upcoming enhancements for functionality and security for RoCE. Various modules of the InfiniBand stack including connection manager, user verbs, core, statistics, resource tracking, device discovery and visibility to applications, net device migration across namespaces at minimum are the key areas to address for supporting RoCE devices in container environment.

Building Efficient Clouds for HPC, Big Data, and Neuroscience Applications over SR-IOV-enabled InfiniBand Clusters

Xiaoyi Lu, The Ohio State University

Single Root I/O Virtualization (SR-IOV) technology has been steadily gaining momentum for high performance interconnects such as InfiniBand. SR-IOV can deliver near-native performance but lacks locality-aware communication support. This talk presented an efficient approach to building HPC clouds based on MVAPICH2 and RDMA-Hadoop with SR-IOV. The talk highlighted high-performance designs of the virtual machine and container aware MVAPICH2 library over SR-IOV enabled HPC Clouds. This talk also presented a high-performance virtual machine migration framework for MPI applications on SR-IOV enabled InfiniBand clouds. The presenter discussed how to leverage the high-performance networking features (e.g., RDMA, SR-IOV) on cloud environments to accelerate data processing through RDMA-Hadoop package. To show the performance benefits of the proposed designs, they co-designed a scalable and distributed tool with MVAPICH2 for statistical evaluation of brain connectomes in the Neuroscience domain, which can run on top of container-based cloud environments with natively utilizing RDMA interconnects and delivering near-native performance.

Non-Contiguous Memory Registration

Tzahi Oved, Mellanox Technologies

Memory registration enables contiguous memory regions to be accessed with RDMA. In this talk, they showed how this could be extended beyond access rights, for describing complex memory layouts. Many HPC applications receive regular structured data, such as a column of a matrix. In this case, the application would typically receive a chunk of data and scatter it by the CPU, or use multiple RDMA writes to transfer each element in-place. Both options introduce significant overhead. By using a memory region that specifies strided access, this overhead could be completely eliminated: the initiator posts a single RDMA write and the target HCA scatters each element into place. Similarly, standard memory regions cannot describe non-contiguous memory allocations, forcing applications to generate remote keys for each buffer. However, by allowing a non-contiguous memory region to span multiple address ranges, an application may scatter remote data with a single remote key. Using non-contiguous memory registration, such memory layouts may be created, accessed, and invalidated using efficient, non-privileged, user-level interfaces.

Dynamically-Connected Transport

Alex Rosenbaum, Mellanox Technologies

Dynamically-Connected (DC) transport is a combination of features from the existing UD and RC transports: DC can send every message to a different destination, like UD does, and is also a reliable transport - supporting RDMA and Atomic operations as RC does. The crux of the transport is dynamically connecting and disconnecting on-the-fly in hardware when changing destinations. As a result, a DC endpoint may communicate with any peer, providing the full RC feature set, and maintain a fixed memory footprint regardless of the size of the network. In this talk, we present the unique characteristics of this new transport, and show how it could be leveraged to reach peek all-to all communication performance. We will review the DC transport objects and their semantics, the Linux upstream DC API and its usage.

T10-DIF offload

Tzahi Oved, Mellanox Technologies

T10-DIF is a standard that defines how to protect the integrity of storage data blocks. Every storage block is proceeded by a Data Integrity Field (DIF). This field contains CRC of the preceding block, the LBA (block number within the storage device) and an application tag. Normally the DIF will be saved in the storage device along with the data block itself, so that in the future it will be used to verify the data integrity.

Modern storage systems and adapters allow creating, verifying and stripping those DIFs while reading and writing data to the storage device, as requested by the user and supported by the OS. The T10-DIF offload RDMA feature brings this capability to the RDMA based storage protocols. Using this feature, RDMA based protocols can request the RDMA device to generate, strip and/or verify DIF while sending or receiving a message. DIF operation is configured in a new Signature Memory-Region. Every memory access using this MR (local or remote) results in DIF operation done on the data as it moves between wire and memory. This session will describe how the configuration and operation of this feature should be done using verbs API.

NVMf Target Offload

Liran Liss, Mellanox Technologies

NVMe is a standard that defines how to access a solid-state storage device over PCI in a very efficient way. It defines how to create and use multiple submission and completion queues between software and the device over which storage operations are carried and completed.

NVMe-over-Fabric is a newer standard that maps NVMe to RDMA to allow remote access to storage devices over an RDMA fabric using the same NVMe language. Since NVMe queues look and act very much like RDMA queues, it is a natural application to bridge between the two. In fact, a couple of software packages today implement an NVMe-over-Fabric to local NVMe target.

The NVMe-oF Target Offload feature is such an implementation that is done in hardware. A supporting RDMA device is configured with the details of the queues of an NVMe device. An incoming client RDMA connection (QP) is then bound to those NVMe queues. From that point on, every IO request arriving over the network from the client is submitted to the respective NVMe queue without any software intervention using PCI peer-to-peer access. This session will describe how the configuration and operation of such feature should be done using verbs.

High-Performance Big Data Analytics with RDMA over NVM and NVMe-SSD

Xiaoyi Lu, The Ohio State University

The convergence of Big Data and HPC has been pushing the innovation of accelerating Big Data analytics and management on modern HPC clusters. Recent studies have shown that the performance of Apache Hadoop, Spark, and Memcached can be significantly improved by leveraging the high performance networking technologies, such as Remote Direct Memory Access (RDMA). Most of these studies are based on `DRAM+RDMA’ schemes. On the other hand, Non-Volatile Memory (NVM)and NVMe-SSD technologies can support RDMA access with low-latency, high-throughput, and persistence on HPC clusters. NVMs and NVMe-SSDs provide the opportunity to build novel high-performance and QoS-aware communication and I/O subsystems for data-intensive applications. In this talk, we proposed new communication and I/O schemes for these data analytics stacks, which are designed with RDMA over NVM and NVMe-SSD. Our studies show that the proposed designs can significantly improve the communication, I/O, and application performance for Big Data analytics and management middleware, such as Hadoop, Spark, Memcached, etc. In addition, we will also discuss how to design QoS-aware schemes in these frameworks with NVMe-SSD.

Comprehensive, Synchronous, High Frequency Measurement of InfiniBand Networks in Production HPC Systems

Michael Aguilar, Sandia National Laboratories

In this presentation, we showed InfiniBand performance information gathered from a large Sandia HPC system, Skybridge. We showed detection of network hot spots that may affect data exchanges for tightly coupled parallel threads. We quantified the overhead cost (application impact) when data is being collected.

At Sandia Labs, we are continuing to develop an InfiniBand fabric switch port sampler that can used to gather remote data from InfiniBand switches. Using coordinated InfiniBand switch and HCA port samplers, a real-time snapshot of InfiniBand traffic can be retrieved from the fabric on a large-scale HPC computing platform. Due to the time-stamped and light-weight data retrieval with LDMS, production job runs can be instrumented to provide research data that can be used to specify computing platforms with improved data performance.

Our implementation of synchronous monitoring of large-scale HPC systems provides insights into how to improve computing performance. Our sampler takes advantage of the OpenFabrics software stack for metric gathering. The OFED stack supports a common inter-operable software stack that provides the inherent ability to gather traffic metrics from selected connection points within a network fabric. We use OFED MAD and UMAD to collect the remote switch port traffic metrics.

The OFA Workshop is extremely valuable to InfiniBand Trade Association members and the fabrics community as a whole with an aim to identify, discuss and overcome the industry’s most significant challenges. We look forward to participating again next year. Videos of each presentation from the OFA Workshop 2018 are now available online on insideHPC.com.

Bill Lee

Author: admin Categories: InfiniBand, OpenFabrics Alliance, RDMA, RoCE Tags:

Enabling Exascale with Co-Design Architecture, Intelligent Networks and InfiniBand Routing

June 14th, 2017

isc

For those working in the High Performance Computing (HPC) industry, achieving Exascale performance has been an ongoing challenge and a significant milestone for some time. Recently, experts have started to take a holistic system-level approach to performance improvements by examining how hardware and software components interact within data centers. This approach, known as co-design architecture, recognizes that the CPU has reached the limits of its scalability, and offers an intelligent network as the new “co-processor” to share the responsibility for handling and accelerating application workloads.

Next week, IBTA representatives will join other industry experts in Frankfurt, Germany for ISC High Performance, an annual event focused on HPC technological developments and applications. In a Birds of a Feather (BoF) session titled A Holistic Approach to Exascale - Co-Design Architecture, Intelligent Networks and InfiniBand Routing, Scott Atchely from Oak Ridge National Laboratory, as well as Richard Graham, Gerald Lotto and Gilad Shainer from member company Mellanox Technologies will discuss the advantages of a co-design architecture in depth. Additionally, the group will cover the role that InfiniBand routers play in reaching Exascale performance and share insights behind these recent HPC developments.

Attending ISC High Performance? The BoF session will be held in the Kontrast room starting at 4:00 p.m. on Monday, June 19. For more information on the BoF session and its participants, visit the event site here.

Bill Lee

RoCE Initiative Launches New Online Product Directory for CIOs and IT Professionals

May 24th, 2017

roce-logo

The RoCE Initiative is excited to announce the launch of the online RoCE Product Directory, the latest technical resource to supplement the IBTA’s RoCE educational program. The new online resource is intended to inform CIOs and enterprise data center architects about their options for deploying RDMA over Converged Ethernet (RoCE) technology within their Ethernet infrastructure.

The directory is comprised of a growing catalogue of RoCE-enabled solutions provided by IBTA members, including Broadcom, Cavium, Inc. and Mellanox Technologies. The new online tool allows users to search by product type and/or brand, connecting them directly to each item’s specific product page. The product directory currently boasts over 65 products and counting that accelerate performance over Ethernet networks while lowering latency.

For more information on the RoCE Product Directory and members currently involved, view the press release here.

Explore the RoCE Product Directory on the RoCE Initiative Product Search page here.

Bill Lee

Author: admin Categories: InfiniBand, RDMA, RoCE Tags:

InfiniBand and RoCE to Make Their Mark at OFA Workshop 2017

March 16th, 2017

openfabriclogo

The OpenFabrics Alliance (OFA) workshop is an annual event devoted to advancing the state of the art in networking. The workshop is known for showcasing a broad range of topics all related to network technology and deployment through an interactive, community-driven event. The comprehensive event includes a rich program made up of more than 50 sessions covering a variety of critical networking topics, which range from current deployments of RDMA to new and advanced network technologies.

To view the full list of abstracts, visit the OFA Workshop 2017 Abstracts and Agenda page.

This year’s workshop program will also feature some notable sessions that showcase the latest developments happening for InfiniBand and RoCE technology. Below are is the collection of OFA Workshop 2017 sessions that we recommend you check out:

Developer Experiences of the First Paravirtual RDMA Provider and Other RDMA Updates
Presented by Adit Ranadive, VMware

VMware’s Paravirtual RDMA (PVRDMA) device is a new NIC in vSphere 6.5 that allows VMs in a cluster to communicate using Remote Direct Memory Access (RDMA), while maintaining latencies and bandwidth close to that of physical hardware. Recently, the PVRDMA driver was accepted as part of the Linux 4.10 kernel and our user-library was added as part of the new rdma-core package.

In this session, we will provide a brief overview of our PVRDMA design and capabilities. Next, we will discuss our development approach and challenges for joint device and driver development. Further, we will highlight our experience for upstreaming the driver and library with the new changes to the core RDMA stack.

We will provide an update on the performance of the PVRDMA device along with upcoming updates to the device capabilities. Finally, we will provide new results on the performance achieved by several HPC applications using VM DirectPath I/O.

This session seeks to engage the audience in discussions on: 1) new RDMA provider development and acceptance, and 2) hardware support for RDMA virtualization.

Experiences with NVMe over Fabrics
Presented by Parav Pandit, Mellanox

NVMe is an interface specification to access non-volatile storage media over PCIe buses. The interface enables software to interact with devices using multiple, asynchronous submission and completion queues, which reside in memory. Consequently, software may leverage the inherent parallelism and low latency of modern NMV devices with minimal overhead. Recently, the NMVe specification has been extended to support remote access over fabrics, such as RDMA and Fibre Channel. Using RDMA, NVMe over Fabrics (NVMe-oF) provides the high BW and low-latency characteristics of NVMe to remote devices. Moreover, these performance traits are delivered with negligible CPU overhead as the bulk of the data transfer is conducted by RDMA.

In this session, we present an overview of NVMe-oF and its implementation in Linux. We point out the main design choices and evaluate NVMe-oF performance for both InfiniBand and RoCE fabrics.

Validating RoCEv2 for Production Deployment in the Cloud Datacenter
Presented by Sowmini Varadhan, Oracle

With the increasing prevalence of ethernet switches and NICs in Data Center Networks, we have been experimenting with the deployment of RDMA over Commodity Ethernet (RoCE) in our DCN. RDMA needs a lossless transport, and, in theory, this can be achieved on ethernet by using priority based PFC (IEEE 802.1qbb) and ECN (IETF RFC 3168).

We describe our experiences in trying to deploy these protocols in a RoCEv2 testbed running @ 100 Gbit/sec consisting of a multi-level CLOS network.

In addition to addressing the documented limitations around PFC/ECN (livelock, pause-frame-storm, memory requirements for supporting multiple priority flows), we also hope to share some of the performance metrics gathered, as well as some feedback on ways to improve the tooling for observability and diagnosability of the system in a vendor-agnostic, interoperable way.

Host Based InfiniBand Network Fabric Monitoring
Presented by Michael Aguilar, Sandia National Laboratories

Synchronized host based InfiniBand network counter monitoring of local connections at 1Hz can provide a reasonable system snapshot understanding of traffic injection/ejection into/from the fabric. This type of monitoring is currently used to enable understanding about the data flow characteristics of applications and inference about congestion based on application performance degradation. It cannot, however, enable identification of where congestion occurs or how well adaptive routing algorithms and policies react to and alleviate it. Without this critical information the fabric remains opaque and congestion management will continue to be largely handled through increases in bandwidth. To reduce fabric opacity, we have extended our host based monitoring to include internal InfiniBand fabric network ports. In this presentation we describe our methodology along with preliminary timing and overhead information. Limitations and their sources are discussed along with proposed solutions, optimizations, and planned future work.

IBTA TWG - Recent Topics in the IBTA, and a Look Ahead
Presented by Bill Magro, Intel on behalf of InfiniBand Trade Association

This talk discusses some recent activities in the IBTA including recent specification updates. It also provides a glimpse into the future for the IBTA.

InfiniBand Virtualization
Presented by Liran Liss, Mellanox on behalf of InfiniBand Trade Association

InfiniBand Virtualization allows a single Channel Adapter to present multiple transport endpoints that share the same physical port. To software, these endpoints are exposed as independent Virtual HCAs (VHCAs), and thus may be assigned to different software entities, such as VMs. VHCAs are visible to Subnet Management, and are managed just like physical HCAs. This session provides an overview of the InfiniBand Virtualization Annex, which was released on November 2016. We will cover the Virtualization model, management, addressing modes, and discuss deployment considerations.

IPoIB Acceleration
Presented by Tzahi Oved, Mellanox

The IPoIB protocol encapsulates IP packets over InfiniBand datagrams. As a direct RDMA Upper Layer Protocol (ULP), IPoIB cannot support HW features that are specific to the IP protocol stack. Nevertheless, RDMA interfaces have been extended to support some of the prominent IP offload features, such as TCP/UDP checksum and TSO. This provided reasonable performance for IPoIB.

However, new network interface features are one of the most active areas of the Linux kernel. Examples include TSS and RSS, tunneling offloads, and XDP. In addition, the basic IP offload features are insufficient to cope with the increasing network bandwidth. Rather than continuously porting IP network interface developments into the RDMA stack, we propose adding abstract network data-path interfaces to RDMA devices.

In order to present a consistent interface to users, the IPoIB ULP continues to represent the network device to the IP stack. The common code also manages the IPoIB control plane, such as resolving path queries and registering to multicast groups. Data path operations are forwarded to devices that implement the new API, or fallback to the standard implementation otherwise. Using the forgoing approach, we show how IPoIB closes the performance gap compared to state-of-the-art Ethernet network interfaces.

Packet Processing Verbs for Ethernet and IPoIB
Presented by Tzahi Oved, Mellanox

As a prominent user-level networking API, the RDMA stack has been extended to support packet processing applications and user-level TCP/IP stacks, initially focusing on Ethernet. This allowed delivering low latency and high message-rate to these applications.

In this talk, we provide an extensive introduction to both current and upcoming packet processing Verbs, such as checksum offloads, TSO, flow steering, and RSS. Next, we describe how these capabilities may also be applied to IPoIB traffic.

In contrast to Ethernet support, which was based on Raw Ethernet QPs that receive unmodified packets from the wire, IPoIB packets are sent over a “virtual wire”, managed by the kernel. Thus, processing selective IP flows from user-space requires coordination with the IPoIB interface.

The Linux SoftRoCE Driver
Presented by Liran Liss, Mellanox

SoftRoCE is a software implementation of the RDMA transport protocol over Ethernet. Thus, any host to conduct RDMA traffic without necessitating a RoCE-capable NIC, allowing RDMA development anywhere.

This session presents the Linux SoftRoCE driver, RXE, which was recently accepted to the 4.9 kernel. In addition, the RXE user-level driver is now part of rdma-core, the consolidated RDMA user-space codebase. RXE is fully interoperable with HW RoCE devices, and may be used for both testing and production. We provide an overview of the RXE driver, detail its configuration, and discuss the current status and remaining challenges in RXE development.

Ubiquitous RoCE
Presented by Alex Shpiner, Mellanox

In recent years, the usage of RDMA in datacenter networks has increased significantly, with RoCE (RDMA over Converged Ethernet) emerging as the canonical approach to deploying RDMA in Ethernet-based datacenters.

Initially, RoCE required a lossless fabric for optimal performance. This is typically achieved by enabling Priority Flow Control (PFC) on Ethernet NICs and switches. The RoCEv2 specification introduced RoCE congestion control, which allows throttling transmission rate in response to congestion. Consequently, packet loss may be minimized and performance is maintained even if the underlying Ethernet network is lossy.

In this talk, we discuss the details of latest developments in the RoCE congestion control. Hardware congestion control reduces the latency of the congestion control loop; it reacts promptly in the face of congestion by throttling the transmission rate quickly and accurately; when congestion is relieved, bandwidth is immediately recovered. The short control loop also prevents network buffers from overfilling in many congestion scenarios.

In addition, fast hardware retransmission complements congestion control in heavy congestion scenarios, by significantly reducing the penalty of packet drops.

Keep an eye out as videos of the OFA Workshop 2017 sessions will be published on both the OFA website and insideHPC. Interested in attending? Registration for the 13th Annual OFA Workshop will be available online and onsite up until the opening day of the event, March 27. Visit the OFA Workshop 2017 Registration page for more information.

Bill Lee

Author: admin Categories: InfiniBand, OpenFabrics Alliance, RoCE Tags:

New InfiniBand Specification Updates Expand Interoperability, Flexibility, and Virtualization Support

November 29th, 2016

ibta_logohorz

Performance demands continue to evolve in High Performance Computing (HPC) and enterprise cloud networks, increasing the need for enhancements to InfiniBand capabilities, support features, and overall interoperability. To address this need, the InfiniBand Trade Association (IBTA) is announcing the public availability of two new InfiniBand Architecture Specification updates - the Volume 2 Release 1.3.1 and a Virtualization Annex to Volume 1 Release 1.3.

The Volume 2 Release 1.3.1 adds flexible performance enhancements to InfiniBand-based networks. With the addition of Forward Error Correction (FEC) upgrades, IT managers can experience both minimal error rates and low latency performance. The new release also enables the InfiniBand subnet manager to optimize signal integrity while maintaining the lowest power possible from the port. Additionally, updates to QSFP28 and CXP28 memory mapping support improved InfiniBand cable management.

This new Volume 2 release also improves upon interoperability and test methodologies for the latest InfiniBand data rates, namely EDR 100 Gb/s and FDR 56 Gb/s. These enhancements are achieved through updated EDR electrical requirements, amended testing methodology for EDR Limiting Active Cables, and FDR interoperability and test specification corrections.

With an aim toward supporting the ever-increasing deployment of virtualized solutions in HPC and enterprise cloud networks, the IBTA also published a new Virtualization Annex to Volume 1 Release 1.3. The Annex extends the InfiniBand specification to address multiple virtual machines connected to a single physical port, which allows subnet managers to recognize each logical endpoint and reduces the burden on the subnet managers as networks leverage virtualization for greater system scalability.

The InfiniBand Architecture Specification Volume 2 Release 1.3.1 and Volume 1 Release 1.3 are available for public download here.

Please contact us at press@infinibandta.org with questions about InfiniBand’s latest updates.

Bill Lee

Incorporate Networking into Hyperconverged Integrated Systems to Gain a Market Advantage

August 22nd, 2016

gartner-logo
The concept of hyperconverged integrated systems (HCIS) emerged as data centers considered new ways to increase resource utilization by reducing infrastructure inefficiencies and complexities. HCIS is primarily a software-defined platform that integrates compute, storage, networking resources. The HCIS market is expected to grow 79 percent to reach almost $2 billion this year, driving it into mainstream use in the next five years, according to Gartner.

Since this market is growing so rapidly, Gartner released an exciting new report, “Use Networking to Differentiate Your Hyperconverged System.” In the report, Gartner advises HCIS vendors to focus on networking to gain competitive market advantage by integrating use-case-specific guidelines and case studies in go-to-market efforts.

According to the report, more than 10 percent of HCIS deployments will suffer from avoidable network-induced performance problems by 2018, up from less than one percent today. HCIS vendors can help address expected challenges and add value for buyers by considering high performance networking protocols, such as InfiniBand and RDMA over Converged Ethernet (RoCE), during the system design stage.

The growing scale of HCIS clusters creates challenges such as expanding workload coverage and diminishing competitive product differentiation. This will force HCIS vendors to alter their product lines and marketing efforts to help their offerings stand out from the rest. Integrating the right networking capabilities will become even more important as a growing number of providers look to differentiate their products. The Gartner report states that by 2018, 60 percent of providers will start to offer integration of networking services, together with compute and storage services, inside of their HCIS products.

Until recently, HCIS vendors have often treated networking simply as a “dumb” interconnect. However, when clusters grow beyond a handful of nodes and higher workloads are introduced, issues begin to arise. This Gartner report stresses that treating the network as “fat dumb pipes” will make it harder to troubleshoot application performance problems from an end-to-end perspective. The report also determines that optimizing the entire communications stack is key to driving latency down and it names InfiniBand and RoCE as important protocols to implement for input/output (I/O)-intensive workloads.

As competition in the HCIS market continues to grow, vendors must change their perception of networking and begin to focus on how to integrate it in order to keep a competitive edge. To learn more about how HCIS professionals can achieve this market advantage, download the full report from the InfiniBand Reports page.

GARTNER is a registered trademark and service mark of Gartner, Inc. and/or its affiliates in the U.S. and internationally, and is used herein with permission. All rights reserved.

Bill Lee

Dive into RDMA’s Impact on NVMe Devices at the 2016 Flash Memory Summit

August 5th, 2016

fms

Next week, storage experts will gather at the 2016 Flash Memory Summit (FMS) in Santa Clara, CA, to discuss the current state of flash memory applications and how these technologies are enabling new designs for many products in the consumer and enterprise markets. This year’s program will include three days packed with sessions, tutorials and forums on a variety of flash storage trends, including new architectures, systems and standards.

NVMe technology, and its impact on enterprise flash applications, is among the major topics that will be discussed at the show. The growing industry demand to unlock flash storage’s full potential by leveraging high performance networking has led to the NVMe community to develop a new standard for fabrics. NVMe over Fabrics (NVMe/F) allows flash storage devices to communicate over RDMA fabrics, such as InfiniBand and RDMA over Converged Ethernet (RoCE), and thereby enabling all flash arrays to overcome existing performance bottlenecks.

Attending FMS 2016?

If you’re attending FMS 2016 and are interested in learning more about the importance of RDMA fabrics for NVMe/F solutions, I recommend the following two educational sessions:

NVMe over Fabrics Panel – Which Transport Is Best?
Tuesday, August 9, 2016 (9:45-10:50 a.m.)

Representatives from the IBTA will join a panel to discuss the value of RDMA interconnects for the NVMe/F standard. Attendees can expect to receive an overview of each RDMA fabric and the benefits they bring to specific applications and workloads. Additionally, the session will cover the promise that NVMe/F has for unleashing the potential performance of NVMe drives via mainstream high performance interconnects.

Beer, Pizza and Chat with the Experts
Tuesday, August 9, 2016 (7-8:30 p.m.)

This informal event encourages attendees to “sit and talk shop” with experts about a diverse set of storage and networking topics. As IBTA’s Marketing Work Group Co-Chair, I will be hosting a table focused on RDMA interconnects. I’d love to meet with you to answer questions about InfiniBand and RoCE and discuss the advantages they provide the flash storage industry.

Additionally, there will be various IBTA member companies exhibiting on the show floor, so stop by their booths to learn about the new InfiniBand and RoCE solutions:

·HPE (#600)

· Keysight Technologies (#810)

· Mellanox Technologies (#138)

· Tektronix (#641)

· University of New Hampshire InterOperability Lab (#719)

For more information on the FMS 2016 program and exhibitors, visit the event website.

Bill Lee

Life in the Fast Lane: InfiniBand Continues to Reign as HPC Interconnect of Choice

July 8th, 2016

top500

TOP500.org recently released its latest account of the world’s most powerful supercomputers and, as with previous reports, InfiniBand leads the way. The 47th edition of the bi-annual list shows that 205 of the fastest commercially available systems are accelerated by InfiniBand and OpenFabrics Software (OFS).

The InfiniBand fabric, with the OFS open source software, is the High Performance Computing (HPC) interconnect of choice because it delivers a distinctive combination of superior performance, efficiency, scalability and low latency. InfiniBand is the only open-standard I/O that provides the capability required to handle supercomputing’s high demand for CPU cycles without time wasted on I/O transactions. With today’s supercomputers pushing nearly 100 petaflops on the LINPACK benchmark, the need for efficient, low latency performance is higher than ever.

High Marks for InfiniBand and OFS

  • InfiniBand and OFS systems outperformed competing technologies in overall efficiency, scoring an 85 percent list average for compute efficiency – with one system even reaching an incredible 99.8 percent.
  • The technologies enable 70 percent of the HPC system segment. This segment includes academic, research and government fields.
  • For supercomputers capable of Petascale performance, the number of InfiniBand and OFS systems grew from 33 to 45.

InfiniBand’s ability to carry multiple traffic types over a single connection makes it ideal for clustering, communications, storage and management. As a result, the interconnect technology is used in thousands of data centers, HPC clusters, storage, and embedded application that scale from two nodes to a single cluster of tens-of-thousands of nodes. Supercomputers powered by OFS reach their highest performance capacity through the speed and efficiency delivered by Remote Direct Memory Access (RDMA). In turn, OFS enables RDMA fabrics, such as InfiniBand, to run applications that require extreme speeds, Petascale-level scalability and utility-class reliability.

Check out the full list at www.top500.org.

Bill Lee

InfiniBand Experts Discuss Latest Trends and Opportunities at OFA Workshop 2016

May 24th, 2016

ofaworkshop2016

Each year, OpenFabrics Software (OFS) users and developers gather at the OpenFabrics Alliance (OFA) Workshop to discuss and tackle the most recent challenges facing the high performance storage and networking industry. OFS is an open-source software that enables maximum application efficiency and performance agnostically over RDMA fabrics, including InfiniBand and RDMA over Converged Ethernet (RoCE). The work of the OFA supports mission critical applications in High Performance Computing (HPC) and enterprise data centers, but is also quickly becoming significant in cloud and hyper-converged markets.

In our previous blog, we showcased an IBTA sponsored session that provided an update on InfiniBand virtualization support. In addition to our virtualization update, there were a handful of other notable sessions that highlighted the latest InfiniBand developments, case studies and tutorials. Below is a collection of notable InfiniBand focused sessions that we recommend you check out:

InfiniBand as Core Network in an Exchange Application
Ralph Barth, Deutsche Börse AG; Joachim Stenzel, Deutsche Börse AG

Group Deutsche Boerse is a global financial service organization covering the entire value chain from trading, market data, clearing, settlement to custody. While reliability has been a fundamental requirement for exchanges since the introduction of electronic trading systems in the 1990s, since about 10 years also low and predictable latency of the entire system has become a major design objective. Both issues have been important architecture considerations, when Deutsche Boerse started to develop an entirely new derivatives trading system T7 for its options market in the US (ISE) in 2008. As the best fit at the time a combination of InfiniBand with IBM® WebSphere® MQ Low Latency Messaging (WLLM) as the messaging solution was determined. Since then the same system has been adopted for EUREX, one of the largest derivatives exchanges in the world, and is now also extended to cover cash markets. The session presents the design of the application and its interdependence with the combination of InfiniBand and WLLM. Also practical experiences with InfiniBand in the last couple of years will be reflected upon.

Download: Slides / Video


Experiences in Writing OFED Software for a New InfiniBand HCA
Knut Omang, Oracle

This talk presents experiences, challenges and opportunities as lead developer in initiating and developing OFED stack support (kernel and user space driver) for Oracles InfiniBand HCA integrated in the new SPARC Sonoma SoC CPU. In addition to the physical HCA function SR/IOV is supported with vHCAs visible to the interconnect as connected to virtual switches. Individual driver instances for the vHCAs maintains page tables set up for the HCAs MMU for memory accessible from the HCA. The HCA is designed to scale to a large number of QPs. For minimal overhead and maximal flexibility, administrative operations such as memory invalidations also use an asynchronous work request model similar to normal InfiniBand traffic.

Download: Slides / Video

Fabrics and Topologies for Directly Attached Parallel File Systems and Storage Networks
Susan Coulter, Los Alamos National Laboratory

InfiniBand fabrics supporting directly attached storage systems are designed to handle unique traffic patterns, and they contain different stress points than other fabrics. These SAN fabrics are often expected to be extensible in order to allow for expansion of existing file systems and addition of new file systems. The character and lifetime of these fabrics is distinct from those of internal compute fabrics, or multi-purpose fabrics. This presentation covers the approach to InfiniBand SAN design and deployment as experienced by the High Performance Computing effort at Los Alamos National Laboratory.

Download: Slides / Video


InfiniBand Topologies and Routing in the Real World
Susan Coulter, Los Alamos National Laboratory; Jesse Martinez, Los Alamos National Laboratory

As with all sophisticated and multifaceted technologies - designing, deploying and maintaining high-speed networks and topologies in a production environment and/or at larger scales can be unwieldy and surprising in their behavior. This presentation illustrates that fact via a case study from an actual fabric deployed at Los Alamos National Laboratory.

Download: Slides / Video


InfiniBand Routers Premier
Mark Bloch, Mellanox Technologies; Liran Liss, Mellanox Technologies

InfiniBand has gone a long way in providing efficient large-scale high performance connectivity. InfiniBand subnets have shown to scale to tens of thousands of nodes, both in raw capacity and in management. As demand for computing capacity increases, future clusters sizes might exceed the number of addressable endpoints in a single IB subnet (around 40K nodes). To accommodate such clusters, a routing layer with the same latencies and bandwidth characteristics as switches is required.

In addition, as data center deployments evolve, it becomes beneficial to consolidate resources across multiple clusters. For example, several compute clusters might require access to a common storage infrastructure. Routers can enable such connectivity while reducing management complexity and isolating intra-subnet faults. The bandwidth capacity to storage may be provisioned as needed.

This session reviews InfiniBand routing operation and how it can be used in the future. Specifically, we will cover topology considerations, subnet management issues, name resolution and addressing, and potential implications for the host software stack and applications.

Download: Slides

Bill Lee

Author: admin Categories: InfiniBand, RDMA Tags: , ,

InfiniBand-based Supercomputer to Power New Discoveries at the Library of Alexandria

May 17th, 2016
“Bibliotheca Alexandrina” by Ting Chen is licensed under CC BY 2.0

“Bibliotheca Alexandrina” by Ting Chen is licensed under CC BY 2.0

Thriving civilizations, both past and present, tend to have one important characteristic in common – a vast, dynamic knowledge base. Whether it be the latest advancements in agriculture, civil engineering or battlefield tactics, technological innovation frequently determined the level and reach of a nation’s influence. One of the most prominent examples of knowledge driven supremacy stems from ancient Egypt’s Library of Alexandria.

Built in the third century BCE, the Library of Alexandria was considered the greatest collection of scholarly works and papers in its era. In addition to gifts from distinguished intellectuals and monarchs, the library built up its massive archive by coping documents and scrolls brought into Alexandria via merchants and traders. Its subsequent destruction a few centuries later was thought to be one of the most significant losses of cultural knowledge in world history.

In 2002, Bibliotheca Alexandrina was constructed to commemorate the library’s remarkable history and lasting notoriety. With a mission “to recapture the spirit of openness and scholarship of the original,” the modern library acts as a global center of knowledge and learning. It contains over a million books in six separate libraries and boasts four museums and 13 academic research centers. Furthermore, Bibliotheca Alexandria acts as a mirrored backup for the Internet Archive, a non-profit digital library that offers free access to millions of books, media and software around the world.

In addition to preserving existing knowledge, the library pursues new insights and understanding as well. Bibliotheca Alexandrina is currently building a new supercomputer with that exact goal in mind. Supercomputing is considered by many to be the standard-bearer of knowledge creation, with many countries committing significant resources to build the world’s most powerful systems (see our blog “Race to Exascale – Nations Vie to Build Fastest Supercomputer”). Supercomputing allows companies, researchers and institutions to process massive data sets to produce useful results in rapid time.

According to a recent announcement from Huawei, the new supercomputer will feature high-density FusionServer X6800 servers powered by high-performance InfiniBand interconnects. The system will be capable of a theoretical peak speed of 118 TFLOPS and a storage capacity of 288 TB. Its design enables an expansion of up to 4.5 PB, ensuring future storage scalability. Once completed, the supercomputer will support a variety of research fields, including bioinformatics, data mining, physics, weather forecasting, resource exploration/extraction and cloud computing.

We look forward to seeing what type of breakthroughs originate from Bibliotheca Alexandrina’s new InfiniBand-based supercomputer. It may even be a discovery that will have the same lasting effect as the original Library of Alexandria, which people will talk about thousands of years from now.

Bill Lee