September 11, 2012
Leading financial analysis solutions provider Murex has selected ClusterVision to supply cluster management software and professional support services for its new HPC cluster in Paris. The combined software and services solution will provide cluster management and operational support for Murex’s existing cluster, which is used to develop risk and asset management solutions for financial organisations worldwide.
PARIS, Sept. 11 — ClusterVision, Europe’s dedicated specialist for HPC cluster solutions, has announced the full-scale implementation of a cluster management solution at the leading French software provider Murex. After the completion of a successful trial, Murex will use Bright Cluster Manager, supplied and supported by ClusterVision, as part of an overall HPC management service for its existing GPU powered compute cluster. The cluster is primarily used as an in-house development and testing platform for Murex’s commercial software applications in the field of financial risk and asset management.
Headquartered in Paris France, and with offices throughout Europe, the USA and Asia-Pacific, Murex is one of Europe’s largest software developers, and one of the world’s leading providers of software and services to the financial sector. Its flagship product, Murex MX-3, is used for risk analysis in financial market trading, and has over 36,000 users at 200 institutions in 65 countries worldwide.
The worldwide financial sector is highly regulated, with risk analysis management solutions, such as those developed by Murex, being an essential requirement of any trading organisation. Financial risk management uses Monte-Carlo based simulation, a complex process of data sampling and stochastic analysis, to determine the uncertainty associated with a given set of financial assets. Risk analysis is both computationally and data intensive, and is one of the most rapidly growing application areas for high performance computing. Murex’s new HPC capability now allows the management of complex financial products in high precision, and in near real time, compared to a previous capability of only computing analytics once or twice a day.
Murex use the compute cluster as a platform for in-house software development and testing. The system comprises a single head node, with redundant power supply and hard disk drives, and 16 IBM iDataplex dx360 M3 server compute nodes. Each server incorporates 2 of Intel’s X5675 6-core, 3.06Ghz clock speed processors and 2 NVIDIA Tesla M2090 accelerators. System interconnect is managed via GigaEthernet and Mellanox QDR infiniband, and storage is provided by IBM’s high performance clustered file system, GPFS. Large volume data analytics applications, such as the financial analysis solutions developed by Murex, are particularly suited to acceleration, a task which is addressed using NVIDIA’s Tesla M2090 GPU processors. The Tesla M2090 GPU has 512 CUDA parallel processing cores, and is capable of accelerating pure CPU performance by up to 10-times.
Finding an effective cluster management solution and on-going professional support were critical challenges in completing the system. In both respects, Murex was recommended by Bright Computing, developers of Bright Cluster Manager, to European cluster specialist ClusterVision. As a relative newcomer to cluster computing, Murex needed cluster management that was easy to install and operate, and that would create a robust reference platform for their own software development process. With a dedicated focus on HPC, ClusterVision represented an ideal partner. Highly experienced in both developing and delivering cluster management solutions, ClusterVision was also able to provide professional support and administration services for Murex’s existing cluster infrastructure, and specialist training for their administrators and users.
After an evaluation of their requirements, ClusterVision supplied Murex with Bright Cluster Manager, initially as a 4-node, 6-month, proof-of-concept demonstration. To ensure the quality of their software development, Murex’s process requires a consistent reference platform for each software release. Once an initial configuration is complete, Bright Cluster Manager is used to capture a complete image of the master node. Subsequent duplication of this image to all other nodes in the system ensures that they are strictly identical, thereby providing the overall reference and consistency needed for each new release.
Following a successful trial period, Murex has now proceeded with a full cluster management implementation, comprising 17 Bright Cluster Manager licences, together with a 3-year provision of associated professional services. Under the service agreement ClusterVision will supply a single-point of contact, first-level support for Murex’s cluster hardware, delivery and support for the Bright Cluster Manager software, and an on-going programme of skills training for their system administrators and end-users. ClusterVision will also assist the team at Murex in their evaluation of the latest release of Bright Cluster Manager, version 6.0, where the Amazon EC2 cloud bursting capabilities are of particular interest.
“IT resources are rare these days and the Bright Computing cluster management suite enabled us to avoid investing a huge number of man days rediscovering what others in research and the industry have already done, or re-developing something internally that we would have had to support for years. The possibility to extend our cluster with EC2 is perfectly suited to our usage since, from time to time, we need to run some benchmarks on huge test cases which exceed our own compute capacity,” says Pierre Spatz, Head of Quantitative Analysis Murex.
“We are extremely pleased to have demonstrated the value of our experience and solution, and to be working with Murex to provide management and on-going support for their growing HPC infrastructure,” said Christopher Huggins, Commercial Director at ClusterVision. “We are seeing a rapid rise in interest in HPC from the financial sector, and our team are currently involved in a number of large scale projects, delivering both turn-key cluster systems and customised service work packages. We are confident that the cluster management and support services which we have put in place at Murex will bring value to their development process, and through their customer relationships will positively impact a wide range of commercial organisations.”
Building on over 25 years of successful presence in capital markets, Murex has developed an unmatched competence in the design and implementation of integrated trading, risk management and processing solutions for top financial institutions, clearing houses, corporations and utilities located across the globe. Our 200 clients range from leading market makers to large-sized or medium-sized buy-side and sell-side institutions. Over 36,000 users rely on MX.3™, the latest Murex platform, and on our strong market knowledge to support their businesses and keep pace with new practices induced by market evolution and regulatory changes. Implementations powered by the MXpress™ approach leverage the wealth of business content accumulated by Murex over the two decades through pre-packaged components of the platform while offering an accelerated process of delivery.
ClusterVision specialises in the design, deployment and operation of High Performance Compute (HPC) clusters. Compute clusters are used by pioneering organisations in academia and industry to run high-intensity computing applications in fields such as scientific research and development, manufacturing, healthcare and finance. By combining cutting-edge hardware and software components with a range of customised professional services, ClusterVision helps its customers create top-quality, efficient and reliable HPC solutions. In addition to systems technologies from leading manufacturers, ClusterVision's solutions typically include a range of HPC software components, such as easy to use cluster provisioning, management and monitoring. ClusterVision offers a full end-to-end portfolio of professional services - from system design, assembly and certification, to operational management, support, and training. With a background in applied scientific research, and practical experience in a wide range of HPC technologies, the ClusterVision team has designed and built some of the largest and most complex computational, storage and database clusters in Europe.
10/30/2013 | Cray, DDN, Mellanox, NetApp, ScaleMP, Supermicro, Xyratex | Creating data is easy… the challenge is getting it to the right place to make use of it. This paper discusses fresh solutions that can directly increase I/O efficiency, and the applications of these solutions to current, and new technology infrastructures.
10/01/2013 | IBM | A new trend is developing in the HPC space that is also affecting enterprise computing productivity with the arrival of “ultra-dense” hyper-scale servers.
Ken Claffey, SVP and General Manager at Xyratex, presents ClusterStor at the Vendor Showdown at ISC13 in Leipzig, Germany.
Join HPCwire Editor Nicole Hemsoth and Dr. David Bader from Georgia Tech as they take center stage on opening night at Atlanta's first Big Data Kick Off Week, filmed in front of a live audience. Nicole and David look at the evolution of HPC, today's big data challenges, discuss real world solutions, and reveal their predictions. Exactly what does the future holds for HPC?