Network performance issues recently raised at IN2P3-CC

Slides:



Advertisements
Similar presentations
Routing Routing in an internetwork is the process of directing the transmission of data across two connected networks. Bridges seem to do this function.
Advertisements

Support for high performance UDP/TCP applications Xing Li
Storage System Integration with High Performance Networks Jon Bakken and Don Petravick FNAL.
Application of GMPLS technology to traffic engineering Shinya Tanaka, Hirokazu Ishimatsu, Takeshi Hashimoto, Shiro Ryu (1), and Shoichiro Asano (2) 1:
Precept 3 Host Configuration 1 Peng Sun. What TCP conn. running? Commands netstat [-n] [-p] [-c] (Linux) lsof -i -P (Mac) ss (newer version of netstat)
© 2008 Cisco Systems, Inc. All rights reserved.Cisco ConfidentialPresentation_ID 1 Chapter 5: Inter-VLAN Routing Routing & Switching.
An Analysis of Bulk Data Movement Patterns in Large-scale Scientific Collaborations W. Wu, P. DeMar, A. Bobyshev Fermilab CHEP 2010, TAIPEI TAIWAN
Maximizing End-to-End Network Performance Thomas Hacker University of Michigan October 5, 2001.
Shadow Configurations: A Network Management Primitive Richard Alimi, Ye Wang, and Y. Richard Yang Laboratory of Networked Systems Yale University February.
Oct. 17/ RoN meetingNetwork tests on SURFnet6 Hans Blom & Paola Grosso AIR group - UvA.
Implementing Standard and Extended Access Control List (ACL) in Cisco Routers.
CISCO NETWORKING ACADEMY Chabot College ELEC Router Introduction.
Lawrence G. Roberts CEO Anagran September 2005 Advances Toward Economic and Efficient Terabit LANs and WANs.
© 2008 Cisco Systems, Inc. All rights reserved.Cisco ConfidentialPresentation_ID 1 Chapter 5: Inter-VLAN Routing Routing And Switching.
KEK Network Qi Fazhi KEK SW L2/L3 Switch for outside connections Central L2/L3 Switch A Netscreen Firewall Super Sinet Router 10GbE 2 x GbE IDS.
Dr. John P. Abraham Professor University of Texas Pan American Internet Routing and Routing Protocols.
IRODS performance test and SRB system at KEK Yoshimi KEK Building data grids with iRODS 27 May 2008.
EGEE-III INFSO-RI Enabling Grids for E-sciencE EGEE and gLite are registered trademarks LHCOPN Operations update Guillaume Cessieux.
Maximizing End-to-End Network Performance Thomas Hacker University of Michigan October 26, 2001.
Data and Computer Communications Chapter 2 – Protocol Architecture, TCP/IP, and Internet-Based Applications 1.
Grid Applications for High Energy Physics and Interoperability Dominique Boutigny CC-IN2P3 June 24, 2006 Centre de Calcul de l’IN2P3 et du DAPNIA.
Thoughts on Firewalls: Topologies, Application Impact, Network Management, Tech Support and more Deke Kassabian, April 2007.
Data transfer over the wide area network with a large round trip time H. Matsunaga, T. Isobe, T. Mashimo, H. Sakamoto, I. Ueda International Center for.
Scavenger performance Cern External Network Division - Caltech Datagrid WP January, 2002.
BNL’s Network diagnostic tool IPERF was used and combined with different strategies to analyze network bandwidth performance such as: -Test with iperf.
2005 © SWITCH PERT – Beyond Fat Pipes Simon Leinen.
Chapter 2 Protocols and the TCP/IP Suite 1 Chapter 2 Protocols and the TCP/IP Suite.
National Center for Atmospheric Research Pittsburgh Supercomputing Center National Center for Supercomputing Applications Web100 Basil Irwin & George Brett.
Connect. Communicate. Collaborate perfSONAR MDM Service for LHC OPN Loukik Kudarimoti DANTE.
Online-Offsite Connectivity Experiments Catalin Meirosu *, Richard Hughes-Jones ** * CERN and Politehnica University of Bucuresti ** University of Manchester.
BNL PDN Enhancements. Perimeter Load Balancers Scaleable Performance Fault Tolerance Server Maintainability User Convenience Perimeter Security.
EGEE-II INFSO-RI Enabling Grids for E-sciencE EGEE and gLite are registered trademarks Update on Network Performance Monitoring.
Network infrastructure at FR-CCIN2P3 Guillaume Cessieux – CCIN2P3 network team Guillaume. cc.in2p3.fr On behalf of CCIN2P3 network team LHCOPN.
LHCOPN: Operations status LHCOPN: Operations status cc.in2p3.fr Network team, FR-CCIN2P3 LHCOPN meeting, Barcelona,
INDIANAUNIVERSITYINDIANAUNIVERSITY Status of FAST TCP and other TCP alternatives John Hicks TransPAC HPCC Engineer Indiana University APAN Meeting – Hawaii.
LHCONE Monitoring Thoughts June 14 th, LHCOPN/LHCONE Meeting Jason Zurawski – Research Liaison.
14-Nov-07 OWAMP (One-Way Latencies) BWCTL (Bandwidth Test Control) Jeff Boote Network Performance Tools BOF-SC07.
10-Jun-05 BWCTL (Bandwidth Test Control) Jeff Boote Network Performance Workshop.
EGEE-III INFSO-RI Enabling Grids for E-sciencE EGEE and gLite are registered trademarks LHCOPN operations Presentation and training.
Hiroyuki Matsunaga (Some materials were provided by Go Iwai) Computing Research Center, KEK Lyon, March
Ch. 23, 25 Q and A (NAT and UDP) Victor Norman IS333 Spring 2015.
Network Monitoring Sebastian Büttrich, NSRC / IT University of Copenhagen Last edit: February 2012, ICTP Trieste
Troubleshooting Ben Fineman,
What is a Protocol A set of definitions and rules defining the method by which data is transferred between two or more entities or systems. The key elements.
Network evolving 2020 and beyond
LHCOPN lambda and fibre routing Episode 4 (the path to resilience…)
New network infrastructure at CCIN2P3
LHC[OPN/ONE]  IPv6  status
Solving Real-World Problems with Wireshark
LHCOPN/LHCONE status report pre-GDB on Networking CERN, Switzerland 10th January 2017
2nd Asia Tier Centre Forum Summary report 4th April 2017 edoardo
Status report on LHC_2: ATLAS computing
Selecting Unicast or Multicast Mode
Instructor Materials Chapter 9: Testing and Troubleshooting
Instructor Materials Chapter 1 Ever Wonder How It Works?
James Casey, IT-GD, CERN CERN, 5th September 2005
The transfer performance of iRODS between CC-IN2P3 and KEK
LHCOPN Operations: Yearly review
Network between CC-IN2P3 and KEK
Monitoring Appliance Status
Networking support (SA2) tasks for EGI
Chapter 5: Inter-VLAN Routing
Deployment & Advanced Regular Testing Strategies
Support for high performance UDP/TCP applications
Routing and Switching Essentials v6.0
Chapter 3: Open Systems Interconnection (OSI) Model
Big-Data around the world
Grid Computing 6th FCPPL Workshop
A tool for locating QoS failures on an Internet path
FTS Issue in Beijing Erming PEI 2010/06/18.
Presentation transcript:

Network performance issues recently raised at IN2P3-CC Guillaume.Cessieux@cc.in2p3.fr 2010-06-17 All units are in bit/sec

LHCOPN related (1/1) 2010-05-31: Performance problem with US-T1-BNL or CH-CERN (software report 40M) Paths are 10G (CERN is direct, BNL is reached through CERN) iperf tests, worst case results achieved both ways: FR-CCIN2P3 ↔ CH-CERN 0.95G on a 1G interface 1.83G on a 2G trunk FR-CCIN2P3 ↔ US-T1-BNL 550M on a 1G interface Not maximal maybe due to some other traffics interfering Far above throughput experienced by software No network issue 19 e-mails, case closed 2010-06-07 for network teams GCX 2010-06-17

Generic IP (1/2) IHEP (CN, Beijing), 2010-05-14 Max bandwidth expected: 90M While iperf achieved: IHEP → IN2P3-CC = 1.58M IN2P3-CC → IHEP = 60M Seems a mix of: Sporadic router overload or traffic-engineering experienced around Beijing’s peering (TEIN3/GÉANT) GÉANT’s ticket stillborn (unable to reproduce…) End hosts’ TCP stacks behaving differently during error recovery Two different hosts, in the same IN2P3-CC’s subnet, tested at the same time can give very different results (Sun Solaris vs Linux SL4) IHEP figured out theirs TCP parameters on hosts were mistakenly changed Reverted back on 2010-06-17 to those from 2010-06-04 Then FTS transfers performance back (~60M both way) 22 e-mails, case closed 2010-06-17 GCX 2010-06-17

Generic IP (2/2) KEK (JP, Tokyo), 2010-02-16 ~1G expected, but iperf reported: IN2P3-CC → KEK = 800M KEK → IN2P3-CC = 300M Good results between IN2P3-CC ↔ ICEPP (University of Tokyo) = ~1G And ICEPP and KEK have a very similar path to IN2P3-CC Thought to be a TCP stack issue Bad behaviour of SL3 TCP stack on high latency networks We figured out no production traffic is exchanged between KEK and IN2P3-CC, so no problem to solve… 17 e-mails, case closed 2010-04-14 Future hosts at KEK will be SL4 or SL5 GCX 2010-06-17

Conclusion Problem often lies on sender’s side Hard to clearly discriminate or not networks Sporadic and temporary issues Iperf tests are good but proof can only be obtained on end-hosts really used Killing generic measurement points on sites... Coordinated efforts required on both sides Too much time spent on administrative issues (get access, open ports etc.) Upcoming Clear process to better manage such regular issues at IN2P3-CC Clear actors, roles, responsibilities and tools Spanning across several teams: support, storage, system and network perfSONAR box at IN2P3-CC Regular bandwidth measurements GCX 2010-06-17