Today Mellanox announced that University of Cambridge has selected Mellanox End-to-End Ethernet interconnect solution including Spectrum SN2700 Ethernet switches, ConnectX-4 Lx NICs and LinkX cables for its OpenStack-based scientific research cloud. This new win has expanded Mellanox’s existing footprint of InfiniBand solution and empowers the UoC to realize its vision of HPC and Cloud convergence through high-speed cloud networks at 25/50/100Gb/s throughput.
Mellanox is seeking an HPC Application Performance Engineer in our Job of the Week. “Mellanox Technologies is looking for a talented engineer to lead datacenter application performance optimization and benchmarking over Mellanox networking products. This individual will primarily work with marketing and engineering to execute low-level and application level benchmarks focused on High Performance Computing (HPC) open source and ISV applications in addition to providing software and hardware optimization recommendations. In addition, this individual will work closely with hardware and software partners, and customers to benchmark Mellanox products under different system configurations and workloads.”
In this video from the 2016 OpenFabrics Workshop, Zili Zheng from LBNL presents: UPC++. “UPC++ is a parallel programming extension for developing C++ applications with the partitioned global address space (PGAS) model. UPC++ has demonstrated excellent performance and scalability with applications and benchmarks such as global seismic tomography, Hartree-Fock, BoxLib AMR framework and more. In this talk, we will give an overview of UPC++ and discuss the opportunities and challenges of leveraging modern network features.”
In this slidecast, Gilad Shainer from Mellanox describes the advantages of InfiniBand and the company’s off-loading network architecture for HPC. “The path to Exascale computing is clearly paved with Co-Design architecture. By using a Co-Design approach, the network infrastructure becomes more intelligent, which reduces the overhead on the CPU and streamlines the process of passing data throughout the network. A smart network is the only way that HPC data centers can deal with the massive demands to scale, to deliver constant performance improvements, and to handle exponential data growth.”
The hpc-ch Forum on Intra- and Inter-Site Networking has posted its Call for Participation. Hosted by the University of Zurich, the event will take place Thursday, May 19, 2016.
“Over the last years the OFA community has shown the potential of using high performance networks (InfiniBand) to boost the performance of virtualized cloud environments, however, the network reconfiguration challenges still continue to exist. In this session we present the work we have been doing on InfiniBand subnet management and routing, in the context of dynamic cloud environments. This work includes, but not limited to, techniques in order to provide better management scalability when virtual machines are live migrating, tenant network isolation in multi-tenant environments, and fast performance-driven network reconfiguration.”
“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 will present 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.”
Today One Stop Systems announced the results of High-Tech Online Magazine Tom’s IT PRO’s review of the 200TB Flash Storage Array SAN (FSA-SAN). OSS’ FSA-SAN provides 200TB of shared Flash RAIDed memory with a choice of the fastest SAN connectivity options including Fiber Channel, Infiniband, and iSCSI.
A design for virtual Ethernet networks over IB is described. The virtual Ethernet networks are implemented as overlays on the IB network. They are managed in a flexible manner using software. Virtual networks can be created, removed, and assigned to servers dynamically using this software. A virtual network can exist entirely on the IB fabric, or it can have an uplink connecting it to physical Ethernet using a gateway. The virtual networks are represented on the servers by virtual network interfaces which can be used with para-virtualized I/O, SRIOV,and non-virtualized I/O. This technology has many uses: communication between applications which are not IB-aware, communication between IB-connected servers and Ethernet-connected servers, and multi-tenancy for cloud environments. It can be used in conjunction with OpenStack, such as for tenant networks.”
In this special guest feature, Ken Strandberg offers this live report from Day 2 of the Lustre User Group meeting in Portland. “Scott Yockel from Harvard University shared how they are deploying Lustre across their massive three data centers up to 90 miles apart with 25 PB of storage, about half of which is Lustre. They’re using Docker containers and employing a backup strategy across the miles of every NFS system, parsing of the entire MDT, and includes 10k directories of small files.”