LHCb on-line / off-line computing. 4

Size: px
Start display at page:

Download "LHCb on-line / off-line computing. 4"

Transcription

1 Off-line computing LHCb on-line / off-line computing Domenico Galli, Bologna INFN CSN Lecce, We plan LHCb-Italy off-line computing resources to be as much centralized as possible Put as much computing power as possible in CNAF Tier- To minimize system administration manpower To optimize resources exploitation Distributed for us means distributed among CNAF and other European Regional Centres Virtual drawback: strong dependence on CNAF resource sharing The improvement following the setup of Tier-3 in major INFN sites for parallel nt-ples analysis should be evaluated later LHCb on-line / off-line computing 2 23 Activities In 23 LHCb-Italy contributed to DC3 (production of MC samples for TDR) 47 Mevt / 6 d 32 Mevt minimum bias; Mevt inclusive b; 5 signal samples, whose size is 5 to kevt 8 Computing centres involved 23 Activities (II) Italian contribution to DC3 has been obtained using limited resources (4kSi2, ie GHz PIII s) Larger contibutions (Karlsruhe D, Imperial College, UK) come from the huges, dinamically allocated, resources of these centres DIRAC, LHCb distributed MC production system, has been used to produce 366 jobs; 85% of them run out of CERN with 92% mean efficiency Italian contribution: 5% (should be 5%) LHCb on-line / off-line computing 3 LHCb on-line / off-line computing 4

2 23 Activities (III) DC3 has also been used to validate LHCb distributed analysis model Distribution to Tier- centres of signal and bg MC samples stored at CERN during production Samples has been pre-reduced based on kinematical or trigger criteria Selection algorithms for specific decay channels (~3) has been executed 23 Activities (IV) To perform high statistics data samples analysis the PVFS distributed file system has been used MB/s aggregate I/O using Base-T Ethernet connection (to be compared with 5 MB/s of a typical Base-T NAS) Events has been classified by means of tagging algorithms LHCb-Italy contributed to implementation of selection algorithms for B decays in 2 charged pions/kaons LHCb on-line / off-line computing 5 LHCb on-line / off-line computing 6 23 Activities (V) Software Roadmap Analysis work by LHCb-Italy has been included in Reoptimized Detector Design and Performance TDR (2 hadrons channel + tagging) 3 LHCb internal notes has been written: CERN-LHCb/23-23: Bologna group, Selection of B/Bs h+h- decays at LHCb ; CERN-LHCb/23-24: Bologna group, CP sensitivity with B/Bs h+h- decays at LHCb CERN-LHCb/23-5: Milano group, LHCb flavour tagging performance LHCb on-line / off-line computing 7 LHCb on-line / off-line computing 8

3 DC4 (April-June 24) Physics Goals Demonstrate performance of HLTs (needed for computing TDR) Large minimum bias sample + signal Improve B/S estimates of optimisation TDR Large bb sample + signal Physics improvements to generators DC4 Computing Goals Main goal: gather information to be used for writing LHCb computing TDR Robustness test of the LHCb software and production system Using software as realistic as possible in terms of performance Test of the LHCb distributed computing model Including distributed analyses Incorporation of the LCG application area software into the LHCb production environment Use of LCG resources as a substantial fraction of the production capacity LHCb on-line / off-line computing 9 LHCb on-line / off-line computing DC4 Production Scenario Generate (Gauss, SIM output): 5 Million events minimum bias 5 Million events inclusive b decays 2 Million exclusive b decays in the channels of interest Digitize (Boole, DIGI output): All events, apply L+L trigger decision Reconstruct (Brunel, DST output): Minimum bias and inclusive b decays passing L and L trigger Entire exclusive b-decay sample Store: SIM+DIGI+DST of all reconstructed events LHCb on-line / off-line computing Goal: Robustness Test of the LHCb Software and Production System First use of the simulation program Gauss based on Geant4 Introduction of the new digitisation program, Boole With HLTEvent as output Robustness of the reconstruction program, Brunel Including any new tuning or other available improvements Not including mis-alignment/calibration Pre-selection of events based on physics criteria (DaVinci) AKA stripping Performed by production system after the reconstruction Producing multiple DST output streams Further development of production tools (Dirac etc) eg integration of stripping eg Book-keeping improvements eg Monitoring improvements LHCb on-line / off-line computing 2

4 Goal: Test of the LHCb Computing Model Distributed data production As in 23, will be run on all available production sites Including LCG Controlled by the production manager at CERN In close collaboration with the LHCb production site managers Distributed data sets CERN: Complete DST (copied from production centres) Master copies of pre-selections (stripped DST) Tier: Complete replica of pre-selections Master copy of DST produced at associated sites Master (unique!) copy of SIM+DIGI produced at associated sites Goal: Incorporation of the LCG Software Gaudi will be updated to: Use POOL (persistency hybrid implementation) mechanism Use certain SEAL (general framework services) services eg Plug-in manager All the applications will use the new Gaudi Should be ~transparent but must be commissioned NB: POOL provides existing functionality of ROOT I/O And more: eg location independent event collections But incompatible with existing TDR data May need to convert it if we want just one data format Distributed analysis LHCb on-line / off-line computing 3 LHCb on-line / off-line computing 4 Needed Resources for DC4 requirement is times what was needed for DC3 Current resource estimates indicate DC4 will last 3 months Assumes that Gauss is twice slower than SICBMC Currently planned for April-June Resources request to Bologna Tier- for DC4 power: 2 ksi2 (5 GHz PIII ) Disk: 5 TB Tape: 5 TB GOAL: use of LCG resources as a substantial fraction of the production capacity We can hope for up to 5% Storage requirement: 6TB at CERN for complete DST 9TB distributed among TIER for locally produced SIM+DIGI+DST up to TB per TIER for pre-selected DSTs LHCb on-line / off-line computing 5 LHCb on-line / off-line computing 6

5 Tier- Grow in Next Years Online Computing LHCb-Italy has been involved in online group to design the L/HLT trigger farm Sezione di Bologna [ksi2] Disk [TB] G Avoni, A Carbone,, U Marconi, G Peco, M Piccinini, V Vagnoni Sezione di Milano T Bellunato, L Carbone, P Dini Sezione di Ferrara A Gianoli Tape [TB] LHCb on-line / off-line computing 7 LHCb on-line / off-line computing 8 Online Computing (II) Lots of changes since the Online TDR abandoned Network Processors included Level- DAQ have now Ethernet from the readout boards destination assignment by TFC (Timing and Fast Control) Main ideas the same large gigabit Ethernet Local Area Network to connect detector sources to destinations simple (push) protocol, no event-manager commodity components wherever possible everything controlled, configured and monitored by ECS (Experimental Control System) LHCb on-line / off-line computing 9 DAQ Architecture Level- Traffic Links 44 khz 55- GB/s Gb Ethernet Level- Traffic HLT Traffic Mixed Traffic Front-end Electronics FE FE FE FE FE FE FE FE FE FE FE FE TRM es Links 88 khz Storage System Readout Network Links GB/s s ~8 s LHCb on-line / off-line computing 2 Multiplexing Layer L-Decision Sorter Farm HLT Traffic 323 Links 4 khz 6 GB/s 29 es 32 Links TFC System

6 Following the Data-Flow Storage System Gb Ethernet Level- Traffic HLT Traffic Mixed Traffic FE FE FE FE FE FE FE FE 2 FE FE FE FE 2 Readout Network 94 Links 7 GB/s 94 s L L Trigger D 2 ~8 s 2 LHCb on-line / off-line computing 2 Front-end Electronics B TRM Sorter L-Decision HLT Yes ΦΚ s Farm L TFCYes L System Yes Design Studies Items under study: Physical farm implementation (choice of cases, cooling, etc) Farm management (bootstrap procedure, monitoring) Subfarm Controllers (event-builders, load-balancing queue) Ethernet es Integration with TFC and ECS System Simulation LHCb-Italy is involved in Farm management, Subfarm Controllers and their communication with Subfarm Nodes LHCb on-line / off-line computing 22 Tests in Bologna To begin the activity in Bologna we started (August 23) from scratch by trying to transfer data through Base-T (gigabit Ethernet on copper cables) from PC to PC and to measure performances As we plan to use an unreliable protocol (RAW Ethernet, RAW IP or UDP) because reliable ones (like TCP, which retransmit datagrams not acknowledged) introduce unpredictable latency, so, together with throughput and latency, we need to benchmark also data loss Tests in Bologna (II) Previous results In IEEE823 standard specifications, for m long cat5e cables, the BER (Bit Error Rates) is said to be < - Previous measures, performed by A Barczyc, B Jost, N Neufeld using Network Processors (not real PCs) and m long cat5e cables showed a BER < -4 Recent measures (presented A Barczyc at Zürich, 8923), performed using PCs gave a frame drop rate O( -6 ) Many data (too much for L!) get lost inside kernel network stack implementation in PCs LHCb on-line / off-line computing 23 LHCb on-line / off-line computing 24

7 Tests in Bologna (III) Transferring data on Base-T Ehernet is not as trivial as it was for Base-TX Ethernet A new bus (PCI-X) and new chipsets (eg Intel E76, 875P) has been designed to support gigabit NIC data flow (PCI bus and old chipsets have not enough bandwidth to support gigabit NIC at gigabit rate) Linux kernel implementation of network stack has been rewritten 2 times since kernel 24 to support gigabit data flow (networking code is 2% of the kernel source) Last modification imply the change of the kernel-to-driver interface (network driver must be rewritten) Standard Linux RedHat 9A setup uses back-compatibility stuff and looses packets No many people are interested in achieving very low packet loss (except for video streaming) Also a DATATAG group is working on packet losses (M Rio, T Kelly, M Goutelle, R Hughes-Jones, JPMartin-Flatin, A map of the networking code in Linux Kernel 242, draft 8, 8 August 23) Tests in Bologna Results Summary Throughput was always higher than expected (957 Mb/s of IP payload measured) while data loss was our main concern We have understood, first (at least) in the LHCb collaboration, how to send IP datagram at gigabit/second rate from Linux to Linux on Base-T Ethernet without datagram loss (4 datagrams lost / 2x datagrams sent) This required: use the appropriate software: NAPI kernel ( 242 ) NAPI-enabled drivers (for Intel e driver, recompilation with a special flag set was needed) kernel parameters tuning (buffer & queue length) Base-T flow control enabled on NIC LHCb on-line / off-line computing 25 LHCb on-line / off-line computing 26 Test-bed 2 x PC with 3 x Base-T interfaces each Motherboard: SuperMicro X5DPL-iGM Dual Pentium IV Xeon 24 GHz, GB ECC RAM Chipset Intel E75 4/533 MHz FSB (front side bus) Bus Controller Hub Intel P64H2 (2 x PCI-X, 64 bit, 66//33 MHz) Ethernet controller Intel 82545EM: x Base-T interface (supports Jumbo Frames) Plugged-in PCI-X Ethernet Card: Intel Pro/ MT Dual Port Server Adapter Ethernet controller Intel 82546EB: 2 x Base-T interfaces (supports Jumbo Frames) Base-T 8 ports switch: HP ProCurve 68 6 Gbps backplane: non-blocking architecture latency: < 25 µs (LIFO 64-byte packets) throughput: 9 million pps (64-byte packets) switching capacity: 6 Gbps Cat 6e cables max 5 MHz (cfr 25 MHz Base-T) Test-bed (II) lhcbcn Base-T switch 2 lhcbcn2 echo > /proc/sys/net/ipv4/conf/all/arp_filter to use only one interface to receive packet owning to a certain network (354, and ) uplink LHCb on-line / off-line computing 27 LHCb on-line / off-line computing 28

8 Test-bed (III) SuperMicro X5DPL-iGM Motherboard (Chipset Intel E75) LHCb on-line / off-line computing 29 Chipset internal bandwidth is granted 64 Gb/s min LHCb on-line / off-line computing 3 Benchmark Software We used 2 benchmark software: Netperf 22p4 UDP_STREAM Self-made basic sender & receiver programs using UDP & RAW IP We discovered a bug in netperf on Linux platform: since Linux calls setsockopt(so_sndbuf) & setsockopt(so_rcvbuf) set the buffer size to twice the requested size, while Linux calls getsockopt(so_sndbuf) & getsockopt(so_rcvbuf) return the actual the buffer size, then when netperf iterate to achieve the requested precision in results, it doubles the buffer size each iteration, using the same variable for both the sistem calls Benchmark Environment Kernel smp GigaEthernet driver: e version 543-k (RedHat 9A) version 526 recompiled with NAPI flag enabled System disconnected from public network Runlevel 3 (X stopped) Daemons stopped (crond, atd, sendmail, etc) Flow control on (on both NICs and switch) Numer of descriptors allocated by the driver rings: 256, 496 IP send buffer size: (x2) Bytes IP receive buffer size: (x2), (x2) Bytes Tx queue length, 6 LHCb on-line / off-line computing 3 LHCb on-line / off-line computing 32

9 First Results Linux RedHat 9A, Kernel 242, Default Setup, no Tuning First benchmark results about datagram loss showed big fluctuations which, in principle, can due to packet queue reset, other process, interrupts, soft_irqs, broadcast network traffic, etc Resulting distribution is multi-modal Mean loss: datagram lost every 2 datagram sent Too much for LHCb L!!! /333 /3 / Lost datagram fraction for datagram runs 26 runs x -3 LHCb on-line / off-line computing 33 5 Mean value: 595 = /963 Flow control on uplink off Tx-ring: 256 descriptors Rx-ring: 256 descriptors Send buffer: B Recv buffer: B Tx queue length: Driver e: 543-k (no NAPI) 6 5 First Results Linux RedHat 9A, Kernel 242, Default Setup, no Tuning (II) / Mean value: 595 = /963 Flow control on uplink off Tx-ring: 256 descriptors Rx-ring: 256 descriptors Send buffer: B Recv buffer: B Tx queue length: Driver e: 543-k (no NAPI) /8696 /7474 /45872 /333 6 / Lost datagram fraction for datagram runs 26 runs x -4 LHCb on-line / off-line computing Mean value: 595 = /963 4 Flow control on uplink off 3 4 Tx-ring: 256 descriptors /3 Rx-ring: 256 descriptors / Send buffer: B Recv buffer: B 2 Tx queue length: Driver e: 543-k (no NAPI) Lost datagram fraction for datagram runs 26 runs x -3 /44643 / Flow control on 5 Mean value: 595 = /963 6 Lost datagram fraction for datagram runs 26 runs x -3 uplink off /42735 Tx-ring: 256 descriptors 2 Rx-ring: 256 descriptors Send buffer: B Recv buffer: B /4652 Tx queue length: 5 Driver e: 543-k (no NAPI) We think that peak behavior is due to kernel queues resets (all queue packets silently dropped when queue is full) Changes in Linux Network Stack Implementation 2 22: netlink, bottom halves, HFC (harware flow control) As few computation as possible while in interrupt context (interrupt disabled) Part of the processing deferred from interrupt handler to bottom halves to be executed at later time (with interrupt enabled) HFC (to prevent interrupt livelock): as the backlog queue is totally filled, interrupt are disable until backlog queue is emptied Bottom halves execution strictly serialized among s; only one packet at a time can enter the system : softnet, softirq softirqs are software thread that replaces bottom halves possible parallelism on SMP machines (NB: back-port): NAPI (new application program interface) interrupt mitigation technology (mixture of interrupt and polling mechanisms) Interrupt livelock Given the interrupt rate coming in, the IP processing thread never gets a chance to remove any packets off the system There are so many interrupts coming into the system such that no useful work is done Packets go all the way to be queued, but are dropped because the backlog queue is full System resourced are abused extensively but no useful work is accomplished LHCb on-line / off-line computing 35 LHCb on-line / off-line computing 36

10 NAPI (New API) NAPI is a interrupt mitigation mechanism constituted by a mixture of interrupt and polling mechanisms: Polling: useful under heavy load introduces more latency under light load abuses the by polling devices that have no packet to offer Interrupts: improve latency under light load make the system vulnerable to livelock as the interrupt load exceed the MLFFR (Maximum Loss Free Forwarding Rate) LHCb on-line / off-line computing 37 cat 5e cable Packet Reception in Linux kernel 249 (softnet) and 242 (NAPI) Softnet (kernel 249) interrupt cat 5e cable DMA engine NIC DMA engine interrupt int handler disable int alloc sk_buff fetch (DMA) netif_rx() enable int rx-ring descriptor point to kernel memory push cpu_raise_softirq() (schedule a softirq) rx-ring descriptor # backlog queue # backlog queue if full do empty it completely (to avoid inerrupt livelock, HFC) softirq handler net_rx_action() kernel thread add poll_list int handler device point to eth disable int pointer eth NIC alloc sk_buff fetch (DMA) kernel memory netif_rx_schedule() softirq handler enable int cpu_raise_softirq() net_rx_action() (schedule a softirq) NAPI (kernel 242) kernel thread LHCb on-line / off-line computing 38 read pop ip_rcv() poll ip_rcv() further processing further processing NAPI (II) Under low load, before the MLFFR is reached, the system converges toward an interrupt driven system: packets/interrupt ratio is lower and latency is reduced Under heavy load, the system takes its time to poll devices registered Interrupts are allowed as fast as the system can process them : packets/interrupt ratio is higher and latency is increased NAPI (III) NAPI changes driver-to-kernel interfaces all network drivers should be rewritten In order to accommodate devices not NAPI-aware, the old interface (backlog queue) is still available for the old drivers (back-compatibility) Backlog queues, when used in back-compatibility mode, are polled just like other devices LHCb on-line / off-line computing 39 LHCb on-line / off-line computing 4

11 True NAPI vs Back-Compatibility Mode NAPI DMA engine interrupt rx-ring descriptor add poll_list read cat 5e int handler device point to eth cable disable int pointer eth poll NIC alloc sk_buff fetch (DMA) kernel memory netif_rx_schedule() softirq handler enable int cpu_raise_softirq() (schedule a softirq) net_rx_action() ip_rcv() kernel thread NAPI kernel with NAPI driver # backlog queue rx-ring descriptor interrupt push # backlog queue cat 5e cable DMA engine NIC add poll_list int handler backlog point to backlog disable int eth alloc sk_buff fetch (DMA) kernel memory netif_rx() softirq handler enable int cpu_raise_softirq() net_rx_action() (schedule a softirq) kernel thread LHCb on-line / off-line computing 4 NAPI kernel with old (not NAPI-aware) driver read ip_rcv() poll further processing further processing The Intel e Driver Even in the last version of e driver (526) NAPI is turned off by default (to allow the usage of the driver also in kernels 249) To enable NAPI, e 526 driver must be recompiled with the option: make CFLAGS_EXTRA=-DCONFIG_E_NAPI LHCb on-line / off-line computing 42 Best Results Maximum trasfer rate (udp 496 byte datagrams): 957 Mb/s Mean datagram lost fraction (@ 957 Mb/s): 2x - (4 datagram lost for 2x 4k-datagrams sent) corresponding to BER 62x -5 (using m cat6e cables) if data loss is totally due to hardware CRC errors LHCb on-line / off-line computing 43 To be Tested to Improve Further kernel 25 fully preemptive (real time) sysenter & sysexit (instead of int x8) for context switching following system calls (3-4 times faster) Asynchronous datagram receiving Jumbo frames Ethernet frames whose MTU (Maximum Transmission Unit) is 9 instead of 5 Less IP datagram fragmentation in packets Kernel Mode Linux ( KML is a technology that enables the execution of ordinary userspace programs inside kernel space Protection-by-software (like in Java bytecode) instead of protection-by-hardware System calls become function calls (32 time faster than int x8, 36 time faster than sysenter/sysexit) LHCb on-line / off-line computing 44

12 Milestones 824 Streaming benchmarks: Maximum streaming throughput and packet loss using UDP, RAW IP and ROW Ethernet with loopback cable Test of switch performance (streaming throughput, latency and packet loss, using standard frames and jumbo frames) Maximum streaming throughput and packet loss using UDP, RAW IP and ROW Ethernet for 2 or 3 simultaneous connections on the same PC Test of event building (receive 2 message stream and send joined messages stream) 224 (Sub Farm Controller) to nodes communication: Definition of -to-nodes communication protocol Definition of queueing and scheduling mechanism First implementation of queueing/scheduling procedures (possibly zero-copy) Milestones (II) OS test (if performance need to be improved) kernel Linux 2553 KML (kernel mode linux) Design and test of bootstrap procedures: Measure of the rate of failure of simultaneous boot of a cluster of PCs, using pxe/dhcp and tftp Test of node switch on/off and powe cycle using ASF Design of bootstrap system (rate nodes/proxy servers/servers, sofware alignment among servers) Definition of requirement for the trigger software: error trapping timeout LHCb on-line / off-line computing 45 LHCb on-line / off-line computing 46

Reliability of datagram transmission on Gigabit Ethernet at full link load

Reliability of datagram transmission on Gigabit Ethernet at full link load Reliability of datagram transmission on Gigabit Ethernet at full link load LHCb Technical Note Issue: 1 Revision: 0 Reference: LPHE 2005-06 / LHCB 2004-030 DAQ Created: 17 th Oct 2003 Last modified: 31

More information

HIGH-SPEED data-transmission on commodity hardware

HIGH-SPEED data-transmission on commodity hardware 810 IEEE TRANSACTIONS ON NUCLEAR SCIENCE, VOL. 53, NO. 3, JUNE 2006 High Rate Packets Transmission on Ethernet LAN Using Commodity Hardware A. Barczyk, D. Bortolotti, A. Carbone, J.-P. Dufey, D. Galli,

More information

L1 and Subsequent Triggers

L1 and Subsequent Triggers April 8, 2003 L1 and Subsequent Triggers Abstract During the last year the scope of the L1 trigger has changed rather drastically compared to the TP. This note aims at summarising the changes, both in

More information

High bandwidth, Long distance. Where is my throughput? Robin Tasker CCLRC, Daresbury Laboratory, UK

High bandwidth, Long distance. Where is my throughput? Robin Tasker CCLRC, Daresbury Laboratory, UK High bandwidth, Long distance. Where is my throughput? Robin Tasker CCLRC, Daresbury Laboratory, UK [r.tasker@dl.ac.uk] DataTAG is a project sponsored by the European Commission - EU Grant IST-2001-32459

More information

Advanced Computer Networks. End Host Optimization

Advanced Computer Networks. End Host Optimization Oriana Riva, Department of Computer Science ETH Zürich 263 3501 00 End Host Optimization Patrick Stuedi Spring Semester 2017 1 Today End-host optimizations: NUMA-aware networking Kernel-bypass Remote Direct

More information

Operating Systems. 17. Sockets. Paul Krzyzanowski. Rutgers University. Spring /6/ Paul Krzyzanowski

Operating Systems. 17. Sockets. Paul Krzyzanowski. Rutgers University. Spring /6/ Paul Krzyzanowski Operating Systems 17. Sockets Paul Krzyzanowski Rutgers University Spring 2015 1 Sockets Dominant API for transport layer connectivity Created at UC Berkeley for 4.2BSD Unix (1983) Design goals Communication

More information

Improving Packet Processing Performance of a Memory- Bounded Application

Improving Packet Processing Performance of a Memory- Bounded Application Improving Packet Processing Performance of a Memory- Bounded Application Jörn Schumacher CERN / University of Paderborn, Germany jorn.schumacher@cern.ch On behalf of the ATLAS FELIX Developer Team LHCb

More information

10GE network tests with UDP. Janusz Szuba European XFEL

10GE network tests with UDP. Janusz Szuba European XFEL 10GE network tests with UDP Janusz Szuba European XFEL Outline 2 Overview of initial DAQ architecture Slice test hardware specification Initial networking test results DAQ software UDP tests Summary 10GE

More information

Lighting the Blue Touchpaper for UK e-science - Closing Conference of ESLEA Project The George Hotel, Edinburgh, UK March, 2007

Lighting the Blue Touchpaper for UK e-science - Closing Conference of ESLEA Project The George Hotel, Edinburgh, UK March, 2007 Working with 1 Gigabit Ethernet 1, The School of Physics and Astronomy, The University of Manchester, Manchester, M13 9PL UK E-mail: R.Hughes-Jones@manchester.ac.uk Stephen Kershaw The School of Physics

More information

2008 JINST 3 S Online System. Chapter System decomposition and architecture. 8.2 Data Acquisition System

2008 JINST 3 S Online System. Chapter System decomposition and architecture. 8.2 Data Acquisition System Chapter 8 Online System The task of the Online system is to ensure the transfer of data from the front-end electronics to permanent storage under known and controlled conditions. This includes not only

More information

Trigger and Data Acquisition at the Large Hadron Collider

Trigger and Data Acquisition at the Large Hadron Collider Trigger and Data Acquisition at the Large Hadron Collider Acknowledgments (again) This overview talk would not exist without the help of many colleagues and all the material available online I wish to

More information

Summary of the LHC Computing Review

Summary of the LHC Computing Review Summary of the LHC Computing Review http://lhc-computing-review-public.web.cern.ch John Harvey CERN/EP May 10 th, 2001 LHCb Collaboration Meeting The Scale Data taking rate : 50,100, 200 Hz (ALICE, ATLAS-CMS,

More information

Open Source Traffic Analyzer

Open Source Traffic Analyzer Open Source Traffic Analyzer Daniel Turull June 2010 Outline 1 Introduction 2 Background study 3 Design 4 Implementation 5 Evaluation 6 Conclusions 7 Demo Outline 1 Introduction 2 Background study 3 Design

More information

FPGAs and Networking

FPGAs and Networking FPGAs and Networking Marc Kelly & Richard Hughes-Jones University of Manchester 12th July 27 1 Overview of Work Looking into the usage of FPGA's to directly connect to Ethernet for DAQ readout purposes.

More information

Message Passing Architecture in Intra-Cluster Communication

Message Passing Architecture in Intra-Cluster Communication CS213 Message Passing Architecture in Intra-Cluster Communication Xiao Zhang Lamxi Bhuyan @cs.ucr.edu February 8, 2004 UC Riverside Slide 1 CS213 Outline 1 Kernel-based Message Passing

More information

Network Design Considerations for Grid Computing

Network Design Considerations for Grid Computing Network Design Considerations for Grid Computing Engineering Systems How Bandwidth, Latency, and Packet Size Impact Grid Job Performance by Erik Burrows, Engineering Systems Analyst, Principal, Broadcom

More information

The CMS Event Builder

The CMS Event Builder The CMS Event Builder Frans Meijers CERN/EP-CMD CMD on behalf of the CMS-DAQ group CHEP03, La Jolla, USA, March 24-28 28 2003 1. Introduction 2. Selected Results from the Technical Design Report R&D programme

More information

19: Networking. Networking Hardware. Mark Handley

19: Networking. Networking Hardware. Mark Handley 19: Networking Mark Handley Networking Hardware Lots of different hardware: Modem byte at a time, FDDI, SONET packet at a time ATM (including some DSL) 53-byte cell at a time Reality is that most networking

More information

Networking for Data Acquisition Systems. Fabrice Le Goff - 14/02/ ISOTDAQ

Networking for Data Acquisition Systems. Fabrice Le Goff - 14/02/ ISOTDAQ Networking for Data Acquisition Systems Fabrice Le Goff - 14/02/2018 - ISOTDAQ Outline Generalities The OSI Model Ethernet and Local Area Networks IP and Routing TCP, UDP and Transport Efficiency Networking

More information

Ethernet transport protocols for FPGA

Ethernet transport protocols for FPGA Ethernet transport protocols for FPGA Wojciech M. Zabołotny Institute of Electronic Systems, Warsaw University of Technology Previous version available at: https://indico.gsi.de/conferencedisplay.py?confid=3080

More information

The CMS Computing Model

The CMS Computing Model The CMS Computing Model Dorian Kcira California Institute of Technology SuperComputing 2009 November 14-20 2009, Portland, OR CERN s Large Hadron Collider 5000+ Physicists/Engineers 300+ Institutes 70+

More information

6.9. Communicating to the Outside World: Cluster Networking

6.9. Communicating to the Outside World: Cluster Networking 6.9 Communicating to the Outside World: Cluster Networking This online section describes the networking hardware and software used to connect the nodes of cluster together. As there are whole books and

More information

LHCb Computing Strategy

LHCb Computing Strategy LHCb Computing Strategy Nick Brook Computing Model 2008 needs Physics software Harnessing the Grid DIRC GNG Experience & Readiness HCP, Elba May 07 1 Dataflow RW data is reconstructed: e.g. Calo. Energy

More information

FAQ. Release rc2

FAQ. Release rc2 FAQ Release 19.02.0-rc2 January 15, 2019 CONTENTS 1 What does EAL: map_all_hugepages(): open failed: Permission denied Cannot init memory mean? 2 2 If I want to change the number of hugepages allocated,

More information

Large Receive Offload implementation in Neterion 10GbE Ethernet driver

Large Receive Offload implementation in Neterion 10GbE Ethernet driver Large Receive Offload implementation in Neterion 10GbE Ethernet driver Leonid Grossman Neterion, Inc. leonid@neterion.com Abstract 1 Introduction The benefits of TSO (Transmit Side Offload) implementation

More information

46PaQ. Dimitris Miras, Saleem Bhatti, Peter Kirstein Networks Research Group Computer Science UCL. 46PaQ AHM 2005 UKLIGHT Workshop, 19 Sep

46PaQ. Dimitris Miras, Saleem Bhatti, Peter Kirstein Networks Research Group Computer Science UCL. 46PaQ AHM 2005 UKLIGHT Workshop, 19 Sep 46PaQ Dimitris Miras, Saleem Bhatti, Peter Kirstein Networks Research Group Computer Science UCL 46PaQ AHM 2005 UKLIGHT Workshop, 19 Sep 2005 1 Today s talk Overview Current Status and Results Future Work

More information

Performance Characteristics on Fast Ethernet and Gigabit networks

Performance Characteristics on Fast Ethernet and Gigabit networks Version 2.5 Traffic Generator and Measurement Tool for IP Networks (IPv4 & IPv6) FTTx, LAN, MAN, WAN, WLAN, WWAN, Mobile, Satellite, PLC, etc Performance Characteristics on Fast Ethernet and Gigabit networks

More information

Solutions for iseries

Solutions for iseries Innovative solutions for Intel server integration Integrated IBM Solutions for iseries xseries The IBM _` iseries family of servers, including the newest member, IBM _` i5, offers two solutions that provide

More information

I/O Systems. Amir H. Payberah. Amirkabir University of Technology (Tehran Polytechnic)

I/O Systems. Amir H. Payberah. Amirkabir University of Technology (Tehran Polytechnic) I/O Systems Amir H. Payberah amir@sics.se Amirkabir University of Technology (Tehran Polytechnic) Amir H. Payberah (Tehran Polytechnic) I/O Systems 1393/9/15 1 / 57 Motivation Amir H. Payberah (Tehran

More information

Using Time Division Multiplexing to support Real-time Networking on Ethernet

Using Time Division Multiplexing to support Real-time Networking on Ethernet Using Time Division Multiplexing to support Real-time Networking on Ethernet Hariprasad Sampathkumar 25 th January 2005 Master s Thesis Defense Committee Dr. Douglas Niehaus, Chair Dr. Jeremiah James,

More information

Performance Characteristics on Fast Ethernet, Gigabit and 10 Gigabits networks

Performance Characteristics on Fast Ethernet, Gigabit and 10 Gigabits networks Versions 2.6 Traffic Generator for IP Networks (IPv4 & IPv6) FTTx, LAN, MAN, WAN, WLAN, WWAN, Mobile, Satellite, PLC, etc. Performance Characteristics on Fast Ethernet, Gigabit and 10 Gigabits networks

More information

LHCb Distributed Conditions Database

LHCb Distributed Conditions Database LHCb Distributed Conditions Database Marco Clemencic E-mail: marco.clemencic@cern.ch Abstract. The LHCb Conditions Database project provides the necessary tools to handle non-event time-varying data. The

More information

Performance Optimisations for HPC workloads. August 2008 Imed Chihi

Performance Optimisations for HPC workloads. August 2008 Imed Chihi Performance Optimisations for HPC workloads August 2008 Imed Chihi Agenda The computing model The assignment problem CPU sets Priorities Disk IO optimisations gettimeofday() Disabling services Memory management

More information

A Fast Ethernet Tester Using FPGAs and Handel-C

A Fast Ethernet Tester Using FPGAs and Handel-C A Fast Ethernet Tester Using FPGAs and Handel-C R. Beuran, R.W. Dobinson, S. Haas, M.J. LeVine, J. Lokier, B. Martin, C. Meirosu Copyright 2000 OPNET Technologies, Inc. The Large Hadron Collider at CERN

More information

Linux IP Networking. Antonio Salueña

Linux IP Networking. Antonio Salueña Linux IP Networking Antonio Salueña Preface We will study linux networking for the following case: Intel x86 architecture IP packets Recent stable linux kernel series 2.4.x 2 Overview

More information

Virtualization, Xen and Denali

Virtualization, Xen and Denali Virtualization, Xen and Denali Susmit Shannigrahi November 9, 2011 Susmit Shannigrahi () Virtualization, Xen and Denali November 9, 2011 1 / 70 Introduction Virtualization is the technology to allow two

More information

Internet data transfer record between CERN and California. Sylvain Ravot (Caltech) Paolo Moroni (CERN)

Internet data transfer record between CERN and California. Sylvain Ravot (Caltech) Paolo Moroni (CERN) Internet data transfer record between CERN and California Sylvain Ravot (Caltech) Paolo Moroni (CERN) Summary Internet2 Land Speed Record Contest New LSR DataTAG project and network configuration Establishing

More information

An FPGA-Based Optical IOH Architecture for Embedded System

An FPGA-Based Optical IOH Architecture for Embedded System An FPGA-Based Optical IOH Architecture for Embedded System Saravana.S Assistant Professor, Bharath University, Chennai 600073, India Abstract Data traffic has tremendously increased and is still increasing

More information

The latency of user-to-user, kernel-to-kernel and interrupt-to-interrupt level communication

The latency of user-to-user, kernel-to-kernel and interrupt-to-interrupt level communication The latency of user-to-user, kernel-to-kernel and interrupt-to-interrupt level communication John Markus Bjørndalen, Otto J. Anshus, Brian Vinter, Tore Larsen Department of Computer Science University

More information

... Application Note AN-531. PCI Express System Interconnect Software Architecture. Notes Introduction. System Architecture.

... Application Note AN-531. PCI Express System Interconnect Software Architecture. Notes Introduction. System Architecture. PCI Express System Interconnect Software Architecture Application Note AN-531 Introduction By Kwok Kong A multi-peer system using a standard-based PCI Express (PCIe ) multi-port switch as the system interconnect

More information

QuickSpecs. HP Z 10GbE Dual Port Module. Models

QuickSpecs. HP Z 10GbE Dual Port Module. Models Overview Models Part Number: 1Ql49AA Introduction The is a 10GBASE-T adapter utilizing the Intel X722 MAC and X557-AT2 PHY pairing to deliver full line-rate performance, utilizing CAT 6A UTP cabling (or

More information

G Robert Grimm New York University

G Robert Grimm New York University G22.3250-001 Receiver Livelock Robert Grimm New York University Altogether Now: The Three Questions What is the problem? What is new or different? What are the contributions and limitations? Motivation

More information

Implementation and Analysis of Large Receive Offload in a Virtualized System

Implementation and Analysis of Large Receive Offload in a Virtualized System Implementation and Analysis of Large Receive Offload in a Virtualized System Takayuki Hatori and Hitoshi Oi The University of Aizu, Aizu Wakamatsu, JAPAN {s1110173,hitoshi}@u-aizu.ac.jp Abstract System

More information

Ethernet Networks for the ATLAS Data Collection System: Emulation and Testing

Ethernet Networks for the ATLAS Data Collection System: Emulation and Testing Ethernet Networks for the ATLAS Data Collection System: Emulation and Testing F. Barnes, R. Beuran, R. W. Dobinson, M. J. LeVine, Member, IEEE, B. Martin, J. Lokier, and C. Meirosu Abstract-- This paper

More information

Computing. DOE Program Review SLAC. Rainer Bartoldus. Breakout Session 3 June BaBar Deputy Computing Coordinator

Computing. DOE Program Review SLAC. Rainer Bartoldus. Breakout Session 3 June BaBar Deputy Computing Coordinator Computing DOE Program Review SLAC Breakout Session 3 June 2004 Rainer Bartoldus BaBar Deputy Computing Coordinator 1 Outline The New Computing Model (CM2) New Kanga/ROOT event store, new Analysis Model,

More information

Update on PRad GEMs, Readout Electronics & DAQ

Update on PRad GEMs, Readout Electronics & DAQ Update on PRad GEMs, Readout Electronics & DAQ Kondo Gnanvo University of Virginia, Charlottesville, VA Outline PRad GEMs update Upgrade of SRS electronics Integration into JLab DAQ system Cosmic tests

More information

Fast packet processing in the cloud. Dániel Géhberger Ericsson Research

Fast packet processing in the cloud. Dániel Géhberger Ericsson Research Fast packet processing in the cloud Dániel Géhberger Ericsson Research Outline Motivation Service chains Hardware related topics, acceleration Virtualization basics Software performance and acceleration

More information

Speeding up Linux TCP/IP with a Fast Packet I/O Framework

Speeding up Linux TCP/IP with a Fast Packet I/O Framework Speeding up Linux TCP/IP with a Fast Packet I/O Framework Michio Honda Advanced Technology Group, NetApp michio@netapp.com With acknowledge to Kenichi Yasukata, Douglas Santry and Lars Eggert 1 Motivation

More information

Chapter 13: I/O Systems. Operating System Concepts 9 th Edition

Chapter 13: I/O Systems. Operating System Concepts 9 th Edition Chapter 13: I/O Systems Silberschatz, Galvin and Gagne 2013 Chapter 13: I/O Systems Overview I/O Hardware Application I/O Interface Kernel I/O Subsystem Transforming I/O Requests to Hardware Operations

More information

FreeBSD Network Performance Tuning

FreeBSD Network Performance Tuning Sucon 2004 Zurich, Switzerland Hendrik Scholz hscholz@raisdorf.net http://www.wormulon.net/ Agenda Motivation Overview Optimization approaches sysctl() tuning Measurement NIC comparision Conclusion Motivation

More information

A generic firmware core to drive the Front-End GBT-SCAs for the LHCb upgrade

A generic firmware core to drive the Front-End GBT-SCAs for the LHCb upgrade A generic firmware core to drive the Front-End GBT-SCAs for the LHCb upgrade F. Alessio 1, C. Caplan, C. Gaspar 1, R. Jacobsson 1, K. Wyllie 1 1 CERN CH-, Switzerland CBPF Rio de Janeiro, Brazil Corresponding

More information

Chapter 4. Routers with Tiny Buffers: Experiments. 4.1 Testbed experiments Setup

Chapter 4. Routers with Tiny Buffers: Experiments. 4.1 Testbed experiments Setup Chapter 4 Routers with Tiny Buffers: Experiments This chapter describes two sets of experiments with tiny buffers in networks: one in a testbed and the other in a real network over the Internet2 1 backbone.

More information

CSE506: Operating Systems CSE 506: Operating Systems

CSE506: Operating Systems CSE 506: Operating Systems CSE 506: Operating Systems Networking & NFS 4 to 7 layer diagram TCP/IP Reality The OSI model is great for undergrad courses TCP/IP (or UDP) is what the majority of world uses Ethernet (or 802.2 or 802.3)

More information

vnetwork Future Direction Howie Xu, VMware R&D November 4, 2008

vnetwork Future Direction Howie Xu, VMware R&D November 4, 2008 vnetwork Future Direction Howie Xu, VMware R&D November 4, 2008 Virtual Datacenter OS from VMware Infrastructure vservices and Cloud vservices Existing New - roadmap Virtual Datacenter OS from VMware Agenda

More information

ntop Users Group Meeting

ntop Users Group Meeting ntop Users Group Meeting PF_RING Tutorial Alfredo Cardigliano Overview Introduction Installation Configuration Tuning Use cases PF_RING Open source packet processing framework for

More information

cs144 Midterm Review Fall 2010

cs144 Midterm Review Fall 2010 cs144 Midterm Review Fall 2010 Administrivia Lab 3 in flight. Due: Thursday, Oct 28 Midterm is this Thursday, Oct 21 (during class) Remember Grading Policy: - Exam grade = max (final, (final + midterm)/2)

More information

SoftRDMA: Rekindling High Performance Software RDMA over Commodity Ethernet

SoftRDMA: Rekindling High Performance Software RDMA over Commodity Ethernet SoftRDMA: Rekindling High Performance Software RDMA over Commodity Ethernet Mao Miao, Fengyuan Ren, Xiaohui Luo, Jing Xie, Qingkai Meng, Wenxue Cheng Dept. of Computer Science and Technology, Tsinghua

More information

Assessing performance in HP LeftHand SANs

Assessing performance in HP LeftHand SANs Assessing performance in HP LeftHand SANs HP LeftHand Starter, Virtualization, and Multi-Site SANs deliver reliable, scalable, and predictable performance White paper Introduction... 2 The advantages of

More information

Optimizing the GigE transfer What follows comes from company Pleora.

Optimizing the GigE transfer What follows comes from company Pleora. Optimizing the GigE transfer What follows comes from company Pleora. Selecting a NIC and Laptop Based on our testing, we recommend Intel NICs. In particular, we recommend the PRO 1000 line of Intel PCI

More information

Part 1: Introduction to device drivers Part 2: Overview of research on device driver reliability Part 3: Device drivers research at ERTOS

Part 1: Introduction to device drivers Part 2: Overview of research on device driver reliability Part 3: Device drivers research at ERTOS Some statistics 70% of OS code is in device s 3,448,000 out of 4,997,000 loc in Linux 2.6.27 A typical Linux laptop runs ~240,000 lines of kernel code, including ~72,000 loc in 36 different device s s

More information

LHCb Computing Status. Andrei Tsaregorodtsev CPPM

LHCb Computing Status. Andrei Tsaregorodtsev CPPM LHCb Computing Status Andrei Tsaregorodtsev CPPM Plan Run II Computing Model Results of the 2015 data processing 2016-2017 outlook Preparing for Run III Conclusions 2 HLT Output Stream Splitting 12.5 khz

More information

BlueGene/L. Computer Science, University of Warwick. Source: IBM

BlueGene/L. Computer Science, University of Warwick. Source: IBM BlueGene/L Source: IBM 1 BlueGene/L networking BlueGene system employs various network types. Central is the torus interconnection network: 3D torus with wrap-around. Each node connects to six neighbours

More information

Data Transfers Between LHC Grid Sites Dorian Kcira

Data Transfers Between LHC Grid Sites Dorian Kcira Data Transfers Between LHC Grid Sites Dorian Kcira dkcira@caltech.edu Caltech High Energy Physics Group hep.caltech.edu/cms CERN Site: LHC and the Experiments Large Hadron Collider 27 km circumference

More information

The Convergence of Storage and Server Virtualization Solarflare Communications, Inc.

The Convergence of Storage and Server Virtualization Solarflare Communications, Inc. The Convergence of Storage and Server Virtualization 2007 Solarflare Communications, Inc. About Solarflare Communications Privately-held, fabless semiconductor company. Founded 2001 Top tier investors:

More information

Tolerating Malicious Drivers in Linux. Silas Boyd-Wickizer and Nickolai Zeldovich

Tolerating Malicious Drivers in Linux. Silas Boyd-Wickizer and Nickolai Zeldovich XXX Tolerating Malicious Drivers in Linux Silas Boyd-Wickizer and Nickolai Zeldovich How could a device driver be malicious? Today's device drivers are highly privileged Write kernel memory, allocate memory,...

More information

Spanish Tier-2. Francisco Matorras (IFCA) Nicanor Colino (CIEMAT) F. Matorras N.Colino, Spain CMS T2,.6 March 2008"

Spanish Tier-2. Francisco Matorras (IFCA) Nicanor Colino (CIEMAT) F. Matorras N.Colino, Spain CMS T2,.6 March 2008 Spanish Tier-2 Francisco Matorras (IFCA) Nicanor Colino (CIEMAT) Introduction Report here the status of the federated T2 for CMS basically corresponding to the budget 2006-2007 concentrate on last year

More information

Got Loss? Get zovn! Daniel Crisan, Robert Birke, Gilles Cressier, Cyriel Minkenberg, and Mitch Gusat. ACM SIGCOMM 2013, August, Hong Kong, China

Got Loss? Get zovn! Daniel Crisan, Robert Birke, Gilles Cressier, Cyriel Minkenberg, and Mitch Gusat. ACM SIGCOMM 2013, August, Hong Kong, China Got Loss? Get zovn! Daniel Crisan, Robert Birke, Gilles Cressier, Cyriel Minkenberg, and Mitch Gusat ACM SIGCOMM 2013, 12-16 August, Hong Kong, China Virtualized Server 1 Application Performance in Virtualized

More information

Work Project Report: Benchmark for 100 Gbps Ethernet network analysis

Work Project Report: Benchmark for 100 Gbps Ethernet network analysis Work Project Report: Benchmark for 100 Gbps Ethernet network analysis CERN Summer Student Programme 2016 Student: Iraklis Moutidis imoutidi@cern.ch Main supervisor: Balazs Voneki balazs.voneki@cern.ch

More information

LHCb Computing Resources: 2018 requests and preview of 2019 requests

LHCb Computing Resources: 2018 requests and preview of 2019 requests LHCb Computing Resources: 2018 requests and preview of 2019 requests LHCb-PUB-2017-009 23/02/2017 LHCb Public Note Issue: 0 Revision: 0 Reference: LHCb-PUB-2017-009 Created: 23 rd February 2017 Last modified:

More information

ROB IN Performance Measurements

ROB IN Performance Measurements ROB IN Performance Measurements I. Mandjavidze CEA Saclay, 91191 Gif-sur-Yvette CEDEX, France ROB Complex Hardware Organisation Mode of Operation ROB Complex Software Organisation Performance Measurements

More information

Networking in a Vertically Scaled World

Networking in a Vertically Scaled World Networking in a Vertically Scaled World David S. Miller Red Hat Inc. LinuxTAG, Berlin, 2008 OUTLINE NETWORK PRINCIPLES MICROPROCESSOR HISTORY IMPLICATIONS FOR NETWORKING LINUX KERNEL HORIZONTAL NETWORK

More information

I/O. Fall Tore Larsen. Including slides from Pål Halvorsen, Tore Larsen, Kai Li, and Andrew S. Tanenbaum)

I/O. Fall Tore Larsen. Including slides from Pål Halvorsen, Tore Larsen, Kai Li, and Andrew S. Tanenbaum) I/O Fall 2011 Tore Larsen Including slides from Pål Halvorsen, Tore Larsen, Kai Li, and Andrew S. Tanenbaum) Big Picture Today we talk about I/O characteristics interconnection devices & controllers (disks

More information

I/O. Fall Tore Larsen. Including slides from Pål Halvorsen, Tore Larsen, Kai Li, and Andrew S. Tanenbaum)

I/O. Fall Tore Larsen. Including slides from Pål Halvorsen, Tore Larsen, Kai Li, and Andrew S. Tanenbaum) I/O Fall 2010 Tore Larsen Including slides from Pål Halvorsen, Tore Larsen, Kai Li, and Andrew S. Tanenbaum) Big Picture Today we talk about I/O characteristics interconnection devices & controllers (disks

More information

Device-Functionality Progression

Device-Functionality Progression Chapter 12: I/O Systems I/O Hardware I/O Hardware Application I/O Interface Kernel I/O Subsystem Transforming I/O Requests to Hardware Operations Incredible variety of I/O devices Common concepts Port

More information

Chapter 12: I/O Systems. I/O Hardware

Chapter 12: I/O Systems. I/O Hardware Chapter 12: I/O Systems I/O Hardware Application I/O Interface Kernel I/O Subsystem Transforming I/O Requests to Hardware Operations I/O Hardware Incredible variety of I/O devices Common concepts Port

More information

Evaluating the Impact of RDMA on Storage I/O over InfiniBand

Evaluating the Impact of RDMA on Storage I/O over InfiniBand Evaluating the Impact of RDMA on Storage I/O over InfiniBand J Liu, DK Panda and M Banikazemi Computer and Information Science IBM T J Watson Research Center The Ohio State University Presentation Outline

More information

ECE 650 Systems Programming & Engineering. Spring 2018

ECE 650 Systems Programming & Engineering. Spring 2018 ECE 650 Systems Programming & Engineering Spring 2018 Networking Transport Layer Tyler Bletsch Duke University Slides are adapted from Brian Rogers (Duke) TCP/IP Model 2 Transport Layer Problem solved:

More information

URDMA: RDMA VERBS OVER DPDK

URDMA: RDMA VERBS OVER DPDK 13 th ANNUAL WORKSHOP 2017 URDMA: RDMA VERBS OVER DPDK Patrick MacArthur, Ph.D. Candidate University of New Hampshire March 28, 2017 ACKNOWLEDGEMENTS urdma was initially developed during an internship

More information

SEDA: An Architecture for Well-Conditioned, Scalable Internet Services

SEDA: An Architecture for Well-Conditioned, Scalable Internet Services SEDA: An Architecture for Well-Conditioned, Scalable Internet Services Matt Welsh, David Culler, and Eric Brewer Computer Science Division University of California, Berkeley Operating Systems Principles

More information

Best Practices for Deploying a Mixed 1Gb/10Gb Ethernet SAN using Dell EqualLogic Storage Arrays

Best Practices for Deploying a Mixed 1Gb/10Gb Ethernet SAN using Dell EqualLogic Storage Arrays Dell EqualLogic Best Practices Series Best Practices for Deploying a Mixed 1Gb/10Gb Ethernet SAN using Dell EqualLogic Storage Arrays A Dell Technical Whitepaper Jerry Daugherty Storage Infrastructure

More information

Network device drivers in Linux

Network device drivers in Linux Network device drivers in Linux Aapo Kalliola Aalto University School of Science Otakaari 1 Espoo, Finland aapo.kalliola@aalto.fi ABSTRACT In this paper we analyze the interfaces, functionality and implementation

More information

Performance Enhancement for IPsec Processing on Multi-Core Systems

Performance Enhancement for IPsec Processing on Multi-Core Systems Performance Enhancement for IPsec Processing on Multi-Core Systems Sandeep Malik Freescale Semiconductor India Pvt. Ltd IDC Noida, India Ravi Malhotra Freescale Semiconductor India Pvt. Ltd IDC Noida,

More information

Implementing Efficient and Scalable Flow Control Schemes in MPI over InfiniBand

Implementing Efficient and Scalable Flow Control Schemes in MPI over InfiniBand Implementing Efficient and Scalable Flow Control Schemes in MPI over InfiniBand Jiuxing Liu and Dhabaleswar K. Panda Computer Science and Engineering The Ohio State University Presentation Outline Introduction

More information

CSCI-GA Operating Systems. Networking. Hubertus Franke

CSCI-GA Operating Systems. Networking. Hubertus Franke CSCI-GA.2250-001 Operating Systems Networking Hubertus Franke frankeh@cs.nyu.edu Source: Ganesh Sittampalam NYU TCP/IP protocol family IP : Internet Protocol UDP : User Datagram Protocol RTP, traceroute

More information

Chapter 13: I/O Systems

Chapter 13: I/O Systems Chapter 13: I/O Systems DM510-14 Chapter 13: I/O Systems I/O Hardware Application I/O Interface Kernel I/O Subsystem Transforming I/O Requests to Hardware Operations STREAMS Performance 13.2 Objectives

More information

EXTENDING AN ASYNCHRONOUS MESSAGING LIBRARY USING AN RDMA-ENABLED INTERCONNECT. Konstantinos Alexopoulos ECE NTUA CSLab

EXTENDING AN ASYNCHRONOUS MESSAGING LIBRARY USING AN RDMA-ENABLED INTERCONNECT. Konstantinos Alexopoulos ECE NTUA CSLab EXTENDING AN ASYNCHRONOUS MESSAGING LIBRARY USING AN RDMA-ENABLED INTERCONNECT Konstantinos Alexopoulos ECE NTUA CSLab MOTIVATION HPC, Multi-node & Heterogeneous Systems Communication with low latency

More information

by I.-C. Lin, Dept. CS, NCTU. Textbook: Operating System Concepts 8ed CHAPTER 13: I/O SYSTEMS

by I.-C. Lin, Dept. CS, NCTU. Textbook: Operating System Concepts 8ed CHAPTER 13: I/O SYSTEMS by I.-C. Lin, Dept. CS, NCTU. Textbook: Operating System Concepts 8ed CHAPTER 13: I/O SYSTEMS Chapter 13: I/O Systems I/O Hardware Application I/O Interface Kernel I/O Subsystem Transforming I/O Requests

More information

Linux Network Tuning Guide for AMD EPYC Processor Based Servers

Linux Network Tuning Guide for AMD EPYC Processor Based Servers Linux Network Tuning Guide for AMD EPYC Processor Application Note Publication # 56224 Revision: 1.00 Issue Date: November 2017 Advanced Micro Devices 2017 Advanced Micro Devices, Inc. All rights reserved.

More information

Chapter 5 Input/Output. I/O Devices

Chapter 5 Input/Output. I/O Devices Chapter 5 Input/Output 5.1 Principles of I/O hardware 5.2 Principles of I/O software 5.3 I/O software layers 5.4 Disks 5.5 Clocks 5.6 Character-oriented terminals 5.7 Graphical user interfaces 5.8 Network

More information

The Google File System

The Google File System The Google File System Sanjay Ghemawat, Howard Gobioff and Shun Tak Leung Google* Shivesh Kumar Sharma fl4164@wayne.edu Fall 2015 004395771 Overview Google file system is a scalable distributed file system

More information

Operating System: Chap13 I/O Systems. National Tsing-Hua University 2016, Fall Semester

Operating System: Chap13 I/O Systems. National Tsing-Hua University 2016, Fall Semester Operating System: Chap13 I/O Systems National Tsing-Hua University 2016, Fall Semester Outline Overview I/O Hardware I/O Methods Kernel I/O Subsystem Performance Application Interface Operating System

More information

Presentation_ID. 2002, Cisco Systems, Inc. All rights reserved.

Presentation_ID. 2002, Cisco Systems, Inc. All rights reserved. 1 Gigabit to the Desktop Session Number 2 Gigabit to the Desktop What we are seeing: Today s driver for Gigabit Ethernet to the Desktop is not a single application but the simultaneous use of multiple

More information

I/O Devices. I/O Management Intro. Sample Data Rates. I/O Device Handling. Categories of I/O Devices (by usage)

I/O Devices. I/O Management Intro. Sample Data Rates. I/O Device Handling. Categories of I/O Devices (by usage) I/O Devices I/O Management Intro Chapter 5 There exists a large variety of I/O devices: Many of them with different properties They seem to require different interfaces to manipulate and manage them We

More information

Benchmarking message queue libraries and network technologies to transport large data volume in

Benchmarking message queue libraries and network technologies to transport large data volume in Benchmarking message queue libraries and network technologies to transport large data volume in the ALICE O 2 system V. Chibante Barroso, U. Fuchs, A. Wegrzynek for the ALICE Collaboration Abstract ALICE

More information

PacketShader: A GPU-Accelerated Software Router

PacketShader: A GPU-Accelerated Software Router PacketShader: A GPU-Accelerated Software Router Sangjin Han In collaboration with: Keon Jang, KyoungSoo Park, Sue Moon Advanced Networking Lab, CS, KAIST Networked and Distributed Computing Systems Lab,

More information

PCI Express System Interconnect Software Architecture for PowerQUICC TM III-based Systems

PCI Express System Interconnect Software Architecture for PowerQUICC TM III-based Systems PCI Express System Interconnect Software Architecture for PowerQUICC TM III-based Systems Application Note AN-573 By Craig Hackney Introduction A multi-peer system using a standard-based PCI Express multi-port

More information

Receive Livelock. Robert Grimm New York University

Receive Livelock. Robert Grimm New York University Receive Livelock Robert Grimm New York University The Three Questions What is the problem? What is new or different? What are the contributions and limitations? Motivation Interrupts work well when I/O

More information

Developing deterministic networking technology for railway applications using TTEthernet software-based end systems

Developing deterministic networking technology for railway applications using TTEthernet software-based end systems Developing deterministic networking technology for railway applications using TTEthernet software-based end systems Project n 100021 Astrit Ademaj, TTTech Computertechnik AG Outline GENESYS requirements

More information

Introduction to Networks and the Internet

Introduction to Networks and the Internet Introduction to Networks and the Internet CMPE 80N Announcements Project 2. Reference page. Library presentation. Internet History video. Spring 2003 Week 7 1 2 Today Internetworking (cont d). Fragmentation.

More information

An Intelligent NIC Design Xin Song

An Intelligent NIC Design Xin Song 2nd International Conference on Advances in Mechanical Engineering and Industrial Informatics (AMEII 2016) An Intelligent NIC Design Xin Song School of Electronic and Information Engineering Tianjin Vocational

More information