The Internet SLAC Les Cottrell 1 SLAC F

The Internet & SLAC Les Cottrell 1, SLAC F F http: //www. slac. stanford. edu /grp/scs/net/talk/internet-connectivity-97/index. htm F Outline of Talk – I. SLAC’s connectivity – II. How is it Working? – III. Why is it like it is? – IV. What’s going on out there on the Internet? – V. What are we (DOE/Esnet, HEP, SLAC) doing? – VII. Summary & Future 9/14/2021 Talk presented at SLAC, July 1997 1

Some Acronyms ARA - Appletalk Remote Access, protocol to connect up remote Macs F ATM - Autonomous Transfer Mode, a high speed network mechanism F DSL - Digital Subscriber Loop, a proposed medium speed (100 s kbps - Mbps) leased line service (phone company answer to cable modems) F ESnet - Energy Sciences network (DOE’s research network, SLAC’s main connection to Internet F ISDN - Integrated Switched Digital Network, new <= 128 kbps digital switched phone service) F POP - Point of Presence, a place where one or more networks have facilities F SLIP/PPP protocols to provide Internet access over a serial line F VPN - Virtual Private Network, a way of tunneling private data over the public Internet F WAN - Wide Area Network 9/14/2021 2 F

Outline F I. SLAC’s Connectivity F II. How is it Working? F III. Why is it like it is? F IV. What’s going on out there on the Internet? F V. What are we (DOE/Esnet, HEP, SLAC) doing? F VII. Summary & Future 9/14/2021 3

Dial in Access http: //www. slac. stanford. edu/comp/net/residential. html F F F F Terminal/emulator dial in – 7 ports, 14. 4 kbps ARA – 16 ports, <=33. 6 kbps, ~340 accounts (85 active/mo) SLIP/PPP thru campus – 14. 4 kbps, need campus account Netcom, $15/mo, nationwide – 28. 8 kbps Wireless via Ricochet ISDN Direct & via ISP – 9 ports, <=128 kbps, in pilot mode ~25 users, production service late summer Following VPN developments 9/14/2021 4

SLAC’s WAN Connectivity 43 Mbps to ESnet ATM cloud (Sprint Oakland POP) F 1. 5 Mps to Caltech/ESnet F 1. 5 Mbps to LBNL/ESnet F 10 Mbps to Stanford F 9/14/2021 5

Outline F I. SLAC’s Connectivity F II. How is it Working? F III. Why is it like it is? F IV. What’s going on out there on the Internet? F V. What are we (DOE/Esnet, HEP, SLAC) doing? F VII. Summary & Future 9/14/2021 6

What is Important to User F We have to optimize the scarcest & therefore most valuable commodity - Time F How long does it take after I hit the button? 9/14/2021 7

Value of Rapid Response Time F Studies in late 70’s early 80 s by Walt Doherty of IBM & others showed the economic value of rapid response time: – – – 0 -0. 4 s 0. 4 -2 s 2 -12 s 12 s-600 s >600 s = High productivity interactive response = Fully interactive regime =Sporadically Interactive regime =Break in contact regime =Batch regime F There is a threshold around 4 -5 s where complaints increase rapidly 9/14/2021 8

Ping Response for Groups of Hosts International little change N. America E improving 210 ms -> 150 ms N. America W improving 140 ms -> 80 ms 9/14/2021 ESnet improving 100 ms -> 50 ms 9

European/Japan Packet Loss to SLAC F Packet loss much more important – loss of packet typically causes 4 -5 s timeout Increase UK-US bandwidth Improve Esnet Internet connect 9/14/2021 RAL: poor to unnacceptable, most others acceptable 10

Quality by Host Group 0. 0 -1% Good, 1 -5% Acceptable, 5 -12% Poor 12 -25% Bad, > 25% Unusable Similar to Internet Weather Report (<6%, <12%, > 12%) 9/14/2021 11

Outline F I. SLAC’s Connectivity F II. How is it Working? F III. Why is it like it is? F IV. What’s going on out there on the Internet? F V. What are we (DOE/Esnet, HEP, SLAC) doing? F VII. Summary & Future 9/14/2021 12

Driving Forces - Hosts 9/14/2021 13

Driving Forces - New Apps F WWW, multimedia, Internet voice, video conferencing – > 60% graphics – <20% HTML 9/14/2021 14

Driving Forces - Penetration US Domains Countries with Internet access 9/14/2021 15

Current Internet Hosts 9/14/2021 16

Challenge - Diversity of Traffic out of FNAL com 14% 9/14/2021 17

Challenge - No single Mgmt for Links 1 RTR-CGB 4. SLAC. Stanford. EDU 2 RTR-DMZ. SLAC. Stanford. EDU 3 ESNET-A-GATEWAY. SLAC. Stanford. EDU 4 pppl-atms. es. net 5 nynap-pppl-atms. es. net 6 192. 157. 69. 11 [Sprint NAP] 7 core 3 -hssi 3 -0. West. Orange. mci. net 8 core 1. West. Orange. mci. net 9 border 2 -fddi-0. West. Orange. mci. net 10 border 2 -hssi 1 -0 -gw. West. Orange. mci. net 11 192. 204. 183. 3 [PREPnet] 12 DEFAULT 1 -GW. UPENN. EDU 13 NISC 8. UPENN. EDU 9/14/2021 18

Challenge F Commercial Internet focussed on staying alive as opposed to research or promoting advanced requirements 9/14/2021 19

More Acronymns F F F F Cal. REN 2 - a California initiative to provide better educational & research networking CHEP 97 - Computing in High Energy Physics meeting in Berlin, April 1997 ESSC - ESnet’s Steering Committee ICFA - International Committee on Future Accelerators Internet 2 - University initiative to provide improved networking between universities NGI - Next Generation Internet, Presidential initiative v. BNS - very high-speed Backbone Network System, a high speed NSF funded backbone network 9/14/2021 20

Outline F I. SLAC’s Connectivity F II. How is it Working? F III. Why is it like it is? F IV. What’s going on out there on the Internet? F V. What are we (DOE/Esnet, HEP, SLAC) doing? F VII. Summary & Future 9/14/2021 21

New Initiatives - California F Cal. REN 2 – joint proposal NSF, UC, Stanford, Caltech … – includes Pac Bell & Cisco – Distributed Giga. POPs in SF & LA, also SD & Sac – Hi speed (622 Mbps) ring around state envisioned 9/14/2021 22

Bay Area F Bay Area Cal. REN 2 Giga. POP nodes: – UCSF – UCB (links to Esnet, Sprint, MCI/v. BNS, UC Davis (state wide) – UCOP – Stanford (links to NASA/NSI, BBN, MCI & statewide ring) 9/14/2021 23

New U. S. Initiatives: v. BNS NSF initiative for interconnecting supercomputer centers for “meritorious applications” F Extended to promote University interconnectivity F 622 Mbps backbone F 9/14/2021 24

New US Initiatives: Internet 2 F Started out (Oct-96) as consortium of ~ 34 major universities – Now there are over 100 u covers 80% of US university sites we monitor – ~$500 K / university over several years, 25% seed – Will use v. BNS as backbone – Giga. POPs in major areas 9/14/2021 25

Next Generation Internet (NGI) F Presidential Initiative – $100 M/yr for 3 years – 100 sites at 100 times bandwidth (1. 5 Mbps => 155 Mbps backbone) – 10 sites at 1000 times bandwidth – DARPA, DOE, NSF, NASA… F Internet 2/NGI/ESnet relationship unclear – can Universities connect to Internet 2 & ESnet? 9/14/2021 26

U. S. International Connections F Only list those of interest to HEP F Moving to colocate US end points at DC POP to improve peering F Discussing CERN<=>Esnet<=>KEK link F STAR-TAP = proposed Int’l Giga. POP at Chicago 9/14/2021 27

Europe: TEN-34 F W. European and some E. European countries interconnect at 4 - 34 Mbps – de, it, ch, uk, gr, nl, pt, at, lu, es, fr, be, hu, sw+dk+no+fi – Several links in production, more by Jul-97 Intra country links generally good F Intra Europe links improving with TEN-34 F Next step TEN-155 F 9/14/2021 28

Asia & FSU F Most connections thru Japan, in general good to acceptable for KEK – US/Esnet/KEK 522 kbps => 1. 5 Mbps – China 64 kbps => 128 kbps (via KEK) – => 128 kbps BINP/Russia Jun-97 F 2 Mbps 9/14/2021 satellite DESY <=> MSU (Moscow) 29

Outline F I. SLAC’s Connectivity F II. How is it Working? F III. Why is it like it is? F IV. What’s going on out there on the Internet? F V. What are we (DOE/ESnet, HEP, SLAC) doing? F VII. Summary & Future 9/14/2021 30

ESSC End-user Connectivity WG F intra-ESnet connectivity good F ESnet <=> University connectivity often bad F ESSC set up WG to look at problem – Ranked top 20 university sites by ER funding – Monitored from SLAC & FNAL – Identified worst (bad (6) to poor (8) performance) – Recommended ESnet look at 6 BAD sites to understand costs of improving, DOE/ESnet will provide money – Typical Frame Relay 1. 5 Mbps connections $13 K/month 9/14/2021 31

F ESnet Peering Improved peering (63=>110 NSPs), examples: – MCI & Sprint to avoid public interconnect swamps – University of California (avoid Sprint) – THEnet at UT Austin – v. BNS East Coast in place since Feb-97 (avoid W Orange MCI) u West Coast May 1997 u Chicago to come u – Hubs at DC, Oakland, San Diego, Chicago – Now carry > 45 K routes 9/14/2021 32

Improved ESnet Internet connection ESnet Peers with Sprint/MCI to avoid MAE-West 9/14/2021 33

UC-Esnet Improved Peering & UCSC 25 ms 9/14/2021 16 ms 34

v. BNS/Esnet Peering & U. Colorado Improved peering between Esnet & v. BNS 9/14/2021 35

ICFA Internet Working Group F Mini-workshop CHEP 97 F Working groups on: monitoring, remote regions, present status, requirements analysis, and the proposal F End 1998 come up with proposal on what to do & why F Next 9/14/2021 meeting Santa Fe, Sep-97 36

Monitoring - Why F “You can’t manage what you can’t measure” F Monitor to set “user” expectations, help with problem detection, get long term trends End-to-end monitoring mainly using ping F Provides response time, packet loss, reachability, unpredictability F Short (trouble shooting) & long term (planning) F Most important metric is packet loss 9/14/2021 37

Monitoring - Who F Many major HEP sites are monitoring endto-end Internet performance to collaborators – several hundred remote sites monitored F Collaborative effort to provide HEP-wide and ESnet wide reports, requested by ICFA, ESnet – Partially funded by DOE FWP involving SLAC, LBL, HEPNRC – Based on SLAC early work (ping based) will complement LBL NIMI work – SLAC, HEPNRC/FNAL, LBL collaboration 9/14/2021 38

Monitoring - How F Plan to coordinate effort, centered on SLAC/HEPNRC code – install common software – distributed architecture – SLAC, HEPNRC Analysis Sites – Umd, RAL, INFN, KEK, ARM, CMU, RMKI, IN 2 P 3, CERN, DESY, TRIUMF, MSU signed up to be Collection Sites – 247 Remote Sites as of 7/7/97 – Reduces network impact of full mesh monitoring 9/14/2021 39

Data Collection & Distribution Architecture F HTTP WWW E. g. HEPNRC Analysis E. g. RAL E. g. SLAC Analysis Collecting Ping Data (via HTTP) Collecting Pings Remote Remote 9/14/2021 40

Results from ~70 Sites in 10 Countries Being Monitored from SLAC FNAL SLAC 9/14/2021 UMd ORNL Monitoring Site ESnet Site N. American Site International Site 41

Putting it all Together 9/14/2021 42

Outline F I. SLAC’s Connectivity F II. How is it Working? F III. Why is it like it is? F IV. What’s going on out there on the Internet? F V. What are we (DOE/Esnet, HEP, SLAC) doing? F VII. Summary & Future 9/14/2021 43

Summary F Driving forces: – Internet user growth 8. 4 M => 28 M US users (15 mos) – Computer power doubling every 12 -18 months – new applications, WWW, Internet phone, VR, Video. . . F Since Apr-95, no single management for planning, trouble reporting etc. F ESnet performance good to acceptable, N. America poor (~6% packet loss avg), International poor (~7% packet loss avg) F Bottlenecks at interchanges 9/14/2021 44

Future F Many separate initiatives: – critical to make sure they interplay well – identify and avoid bottlenecks – understand guide impact for HEP F Criticality of Internet to HEP collaborations means HEP should increase efforts in this area: – keep tuned in, understand issues – monitor end-to-end performance – work with other research and higher education users 9/14/2021 45

F SLAC Networking F http: //www. slac. stanford. edu/comp/net. html F SLAC F WAN Monitoring Page, lots of pointers http: //www. slac. stanford. edu/comp/net/wan-mon. html F ESnet: http: //www. es. net/ F v. BNS: http: //www. vbns. net/ F Internet 2: http: //www. internet 2. edu/ F NGI: http: //www. hpcc. gov/ngi-concept-08 Apr 97/ F TEN-34: http: //www. scimitar. terena. nl/projects/ten-34/ F ICFA Workshop on HEP & the Internet: F http: //www. slac. stanford. edu/xorg/icfa/chep 97/wg. html 9/14/2021 46