DPDK for ultra low latency applications - Muhammad Ahmad & Ali Rizvi, eMumba Inc.
DPDK for ultra low latency applications - Muhammad Ahmad & Ali Rizvi, eMumba Inc.
Speakers: Muhammad Ahmad, Ali Rizvi
DPDK is the go-to off the shelf, stable and reliable solution for data planes and switching applications globally. It is widely used to accelerate packet processing in various verticals, focusing more on throughput while providing decent latency.
In this presentation, we look at how to use DPDK to provide a network stack solution for ultra-low latency (ULL) applications in the world of algorithmic trading. We examine out of the box latency performance from DPDK. Next, we show how, through systematic tuning and benchmarking, we were able to reduce round trip time (RTT) latency. This involved configuring DPDK in scalar mode, pre-allocating mbuffs by enabling RX bulk allocation and using optimized versions of functions by enabling intrinsics. We used an open source FreeBSD network stack on top of DPDK and modified it in a way that favors low latency (burst_size=1, timeout=0). For low latency use cases, it is necessary that there are no context switches and data shared between the cores, so we used rte_flow to direct packets to specific cores. These optimizations enabled us to process the packets at wire speed and reduce latency by fivefold over the pre-tuning results. For benchmarking at these aggressively low latency levels we built a testbed with commodity hardware providing 7 nanosecond timestamp granularity. We replicated the STAC-T1 test which is a widely accepted latency benchmark in the electronic trading industry.
We also compare the results we achieved with DPDK against those we achieved with OpenOnload TCPDirect, the kernel bypass solution from Solarflare. We conclude with some thoughts on upstream contributions for enabling ULL use cases.
Видео DPDK for ultra low latency applications - Muhammad Ahmad & Ali Rizvi, eMumba Inc. канала DPDK Project
Speakers: Muhammad Ahmad, Ali Rizvi
DPDK is the go-to off the shelf, stable and reliable solution for data planes and switching applications globally. It is widely used to accelerate packet processing in various verticals, focusing more on throughput while providing decent latency.
In this presentation, we look at how to use DPDK to provide a network stack solution for ultra-low latency (ULL) applications in the world of algorithmic trading. We examine out of the box latency performance from DPDK. Next, we show how, through systematic tuning and benchmarking, we were able to reduce round trip time (RTT) latency. This involved configuring DPDK in scalar mode, pre-allocating mbuffs by enabling RX bulk allocation and using optimized versions of functions by enabling intrinsics. We used an open source FreeBSD network stack on top of DPDK and modified it in a way that favors low latency (burst_size=1, timeout=0). For low latency use cases, it is necessary that there are no context switches and data shared between the cores, so we used rte_flow to direct packets to specific cores. These optimizations enabled us to process the packets at wire speed and reduce latency by fivefold over the pre-tuning results. For benchmarking at these aggressively low latency levels we built a testbed with commodity hardware providing 7 nanosecond timestamp granularity. We replicated the STAC-T1 test which is a widely accepted latency benchmark in the electronic trading industry.
We also compare the results we achieved with DPDK against those we achieved with OpenOnload TCPDirect, the kernel bypass solution from Solarflare. We conclude with some thoughts on upstream contributions for enabling ULL use cases.
Видео DPDK for ultra low latency applications - Muhammad Ahmad & Ali Rizvi, eMumba Inc. канала DPDK Project
Показать
Комментарии отсутствуют
Информация о видео
Другие видео канала
DPDK's best kept secret – Micro-benchmark performance testsSolarflare - What is Kernel Bypass, And Why Is It An Important Concept?FPGAs and low latency trading - Williston Hayes - Optiver - FPL2020DPDK, VPP and pfSense 3.0Real Life Financial Network Design - Multicast - BGP - EIGRP - Latency Design ConsiderationsA Low-Latency Library in FPGA Hardware for High-Frequency TradingWhat is a VLAN? Virtual Local Area NetworksCore C++ 2019 :: Nimrod Sapir :: High Frequency Trading and Ultra Low Latency development techniquesBuilding Scalable Data Centers: BGP is the Better IGPThe brief case for User-space Network Stacks (DPDK and friends) The case for DPDK, FD.io VPP, Snabb…Understanding the Performance of DPDK as a Computer ArchitectKernel-bypass networking for fun and profitNeed for Speed - Using DPDK and SR-IOVNetwork Virtualization & Packet Processing on Intel Architecture with Intel DPDKOpen vSwitch DPDK Acceleration Using HW ClassificationNokia - Combining Neutron, DPDK, Ironic and SRIOV for seamless hAdvanced API For Rules Management - RTE Flow Next Evolution - Ori Kam, NVIDIALow Latency Market DataServerless DPDK – How SmartNIC resident DPDK Accelerates Packet Processing1.5: DPDK Introduction