Grid Director™ 4036
Scaling-Out Data Centers with 40Gb/s InfiniBand
The Grid Director 4036 is a high performance, low latency and fully non-blocking InfiniBand switch for high performance clusters. Delivering 2.88Tb/s of non-blocking bandwidth with 100 nanoseconds of latency (port-to-port), I/O bottlenecks are removed making applications operate at maximum efficiency. The efficient Grid Director 4036's smart design makes it easy to build clusters that can scale-out to thousands of nodes.
- 36 40Gb/s ports in a 1U switch with up to 2.88 Tb/s of available bandwidth and ultra-low latency of 100 nanoseconds
- Removes I/O bottlenecks for extreme application performance
- Unlimited scalability across application, database and storage servers
- Simple and fast device management
- Support for longer and more varied cable options
- Fully managed by Unified Fabric Manager™ (UFM) Software
- Fast fabric bring up and advanced routing algorithms
Unified Fabric Manager™ Software for Data Center Management
Scale-Out Fabric Management Software
Mellanox’s Unified Fabric Manager™ (UFM™) is a powerful platform for managing scale-out Ethernet and InfiniBand computing environments. UFM enables data center operators to efficiently provision, monitor and operate the modern data center fabric. UFM boosts application performance and ensures that the fabric is up and running at all times.
UFM uses an innovative application-centric approach to bridge the gap between servers, applications and fabric elements. UFM's revolutionary fabric model allows users to manage fabrics as a set of business related entities such as time critical applications or services. UFM’s management infrastructure enables fabric monitoring and performance optimization on the application-logical level rather than just at the individual port or device level.
- Advantages of the UFM approach include:
- Improved visibility into fabric performance and potential bottlenecks
- Improved performance due to application-centric optimizations
- Quicker troubleshooting time due to advanced event management
- Efficient management of dynamic and multi-tenant environments
- Higher utilization of fabric resources
Granular Network Visibility Enabling Total Control
UFM includes an advanced granular monitoring engine that provides real-time access to fabric data. The software also provides a unique congestion tracking feature that quickly identifies traffic bottlenecks and congestion events spreading over the fabric. This feature enables more accurate problem identification and quicker resolution.
UFM is aware of virtual servers and treats them similar to physical servers. The fabric policy engine enables connectivity and fabric policy to virtual servers as well. Upon VM migration, UFM keeps the connectivity and policy intact – providing a seamless migration environment for the application.
For HPC deployments UFM optimizes routing algorithms by taking into consideration the fabric topology and active applications and their characteristics. In addition, UFM enables segmentation of the fabric into isolated partitions, increasing traffic security and predictable application performance.
Robust, Production-Ready Solution
Fabric-wide maintenance tasks are performed from a central location and improve operational efficiency and control, and group operations are enabled via a single mouse click.
Failovers are handled seamlessly and are transparent to both the user and the applications running on the fabric. This significantly lowers downtime and makes UFM along with Mellanox’s products the ultimate management tool for the most demanding data center environments.