Gilad Shainer, VP of Marketing Dec 2013 Interconnect Your Future

© 2013 Mellanox Technologies 2 - Mellanox Confidential - Leading Supplier of End-to-End Interconnect Solutions MXM Mellanox Messaging Acceleration MXM Mellanox Messaging Acceleration FCA Fabric Collectives Acceleration FCA Fabric Collectives Acceleration Management UFM Unified Fabric Management UFM Unified Fabric Management Storage and Data VSA Storage Accelerator (iSCSI) VSA Storage Accelerator (iSCSI) UDA Unstructured Data Accelerator UDA Unstructured Data Accelerator Comprehensive End-to-End Software Accelerators and Managment Host/Fabric SoftwareICsSwitches/GatewaysAdapter CardsCables/Modules Comprehensive End-to-End InfiniBand and Ethernet Portfolio Metro / WAN

© 2013 Mellanox Technologies 3 - Mellanox Confidential - Mellanox InfiniBand Paves the Road to Exascale Computing Accelerating Half of the World’s Petascale Systems Mellanox Connected Petascale System Examples Accelerating Half of the World’s Petascale Systems Mellanox Connected Petascale System Examples

© 2013 Mellanox Technologies 4 - Mellanox Confidential - TOP500 InfiniBand Accelerated Petascale Capable Machines  Mellanox FDR InfiniBand systems Tripled from Nov’12 to Nov’13 Accelerating 63% of the Petaflop capable systems (12 systems out of 19)

© 2013 Mellanox Technologies 5 - Mellanox Confidential - FDR InfiniBand Delivers Highest Return on Investment Higher is better Source: HPC Advisory Council

© 2013 Mellanox Technologies 6 - Mellanox Confidential - Architectural Foundation for Exascale Computing Connect-IB

© 2013 Mellanox Technologies 7 - Mellanox Confidential - Mellanox Connect-IB The World’s Fastest Adapter  The 7 th generation of Mellanox interconnect adapters  World’s first 100Gb/s interconnect adapter (dual-port FDR 56Gb/s InfiniBand)  Delivers 137 million messages per second – 4X higher than competition  Support the new innovative InfiniBand scalable transport – Dynamically Connected

© 2013 Mellanox Technologies 8 - Mellanox Confidential - Connect-IB Provides Highest Interconnect Throughput Source: Prof. DK Panda Connect-IB FDR (Dual port) ConnectX-3 FDR ConnectX-2 QDR Competition (InfiniBand) Connect-IB FDR (Dual port) ConnectX-3 FDR ConnectX-2 QDR Competition (InfiniBand) Higher is Better Gain Your Performance Leadership With Connect-IB Adapters

© 2013 Mellanox Technologies 9 - Mellanox Confidential - Connect-IB Delivers Highest Application Performance 200% Higher Performance Versus Competition, with Only 32-nodes Performance Gap Increases with Cluster Size

© 2013 Mellanox Technologies 10 - Mellanox Confidential - Dynamically Connected Transport Advantages

© 2013 Mellanox Technologies 11 - Mellanox Confidential - Accelerations for Parallel Programs Mellanox ScalableHPC

© 2013 Mellanox Technologies 12 - Mellanox Confidential - Mellanox ScalableHPC Accelerate Parallel Applications InfiniBand Verbs API MXM Reliable Messaging Optimized for Mellanox HCA Hybrid Transport Mechanism Efficient Memory Registration Receive Side Tag Matching FCA Topology Aware Collective Optimization Hardware Multicast Separate Virtual Fabric for Collectives CORE-Direct Hardware Offload Memory P1 Memory P2 Memory P3 MPI Memory P1P2P3 SHMEM Logical Shared Memory Memory P1P2P3 PGAS Memory Logical Shared Memory

© 2013 Mellanox Technologies 13 - Mellanox Confidential - Mellanox FCA Collective Scalability

© 2013 Mellanox Technologies 14 - Mellanox Confidential - Nonblocking Alltoall (Overlap-Wait) Benchmark CoreDirect Offload allows Alltoall benchmark with almost 100% compute

© 2013 Mellanox Technologies 15 - Mellanox Confidential - Accelerator and GPU Offloads

© 2013 Mellanox Technologies 16 - Mellanox Confidential - GPUDirect 1.0 CPU GPU Chip set GPUMemory InfiniBand System Memory CPU GPU Chip set GPUMemory InfiniBand System Memory Transmit Receive CPU GPU Chip set GPUMemory InfiniBand System Memory 1 1 CPU GPU Chip set GPUMemory InfiniBand System Memory 1 1 Non GPUDirect GPUDirect 1.0

© 2013 Mellanox Technologies 17 - Mellanox Confidential - GPUDirect RDMA Transmit Receive CPU GPU Chip set GPUMemory InfiniBand System Memory 1 1 CPU GPU Chip set GPUMemory InfiniBand System Memory 1 1 GPUDirect RDMA CPU GPU Chip set GPUMemory InfiniBand System Memory 1 1 CPU GPU Chip set GPUMemory InfiniBand System Memory 1 1 GPUDirect 1.0

© 2013 Mellanox Technologies 18 - Mellanox Confidential - GPU-GPU Internode MPI Latency Lower is Better 67 % 5.49 usec Performance of MVAPICH2 with GPUDirect RDMA 67% Lower Latency 5X GPU-GPU Internode MPI Bandwidth Higher is Better 5X Increase in Throughput Source: Prof. DK Panda

© 2013 Mellanox Technologies 19 - Mellanox Confidential - Execution Time of HSG (Heisenberg Spin Glass) Application with 2 GPU Nodes Source: Prof. DK Panda Performance of MVAPICH2 with GPU-Direct-RDMA Problem Size

© 2013 Mellanox Technologies 20 - Mellanox Confidential - Roadmap

© 2013 Mellanox Technologies 21 - Mellanox Confidential - Technology Roadmap – One-Generation Lead over the Competition Gbs 40Gbs 56Gbs 100Gbs “Roadrunner” Mellanox Connected “Roadrunner” Mellanox Connected 1 st 3 rd TOP Virginia Tech (Apple) TOP Virginia Tech (Apple) Gbs Mega Supercomputers TerascalePetascaleExascale Mellanox

© 2013 Mellanox Technologies 22 - Mellanox Confidential - Paving The Road for 100Gb/s and Beyond Recent Acquisitions are Part of Mellanox’s Strategy to Make 100Gb/s Deployments as Easy as 10Gb/s Copper (Passive, Active)Optical Cables (VCSEL)Silicon Photonics

© 2013 Mellanox Technologies 23 - Mellanox Confidential - The Only Provider of End-to-End 40/56Gb/s Solutions From Data Center to Metro and WAN X86, ARM and Power based Compute and Storage Platforms The Interconnect Provider For 10Gb/s and Beyond Host/Fabric SoftwareICsSwitches/GatewaysAdapter CardsCables/Modules Comprehensive End-to-End InfiniBand and Ethernet Portfolio Metro / WAN

Thank You