SlideShare a Scribd company logo
1 of 14
Equity Trading: Performance, Latency & Throughput

Prepared for the 2012 Exactpro ExTent Conference, Kostroma, Russia.
11° April 2012
                                                                              Author:
                                                                      Phil Penhaligan
Agenda

  • Low Latency
     –   What are we measuring?
     –   How Fast Is Fast Enough?
     –   Who really needs it?
     –   What’s the next best thing?
     –   The Technology
          • How did we get here?
          • What next?

  • Performance & Capacity Management
     –   The LSE/TQ KPIs
     –   System behaviour under extreme conditions




                                                     2
Low Latency: What are we measuring?
  • 3 distinct attributes
      –   Measured from just outside LSE/TQ firewalls
      –   Excludes transmission time to/from external client/venue
      –   Excludes latency of client/venue’s own system(s)
      –   Excludes network packet retransmissions due to slow consumers


          1. Order latency
              • Time from Order receipt to order acknowledgment
              • Commonly referred to as Order:Ack
          2. Market Data latency
              • Time from Order acknowledgement to public broadcast
                Commonly referred to as Ack:Tick
          3. Reference price latency (Turquoise Dark Book only)
              • Time from Tick receipt to matching engine usage


                                                                          3
Low Latency: What are we measuring?
  • 2 key metrics
     –   Average
           • Mathematical mean
     –   Consistency
           • Max 99.9 percentile
                – Discard the worst 0.1% of orders and then take the maximum

  • Production Systems
     –   Continuous real-time capture
     –   Historical data stored in database for statistical analysis and reporting

  • QA Systems
     –   New software releases evaluated during technical tests
     –   To confirm/verify expected latency improvements
     –   To detect unexpected latency regressions

                                                                                     4
Low Latency: Example – Turquoise Order:Ack

Latency Distribution        Average and Consistency

                              Metric                  uS

                              Average 100%        108


                              Average 99.9%       103


                              Max 99.9%           381


                              Max 99              191




                                                           5
Low Latency: How Fast Is Fast Enough?
• Customer feedback
   –   LSE/TQ averages are already fast enough – better Consistency is most important!
• Our Focus
   –   Since the initial deployment of Millennium (Oct 2009 Turquoise and Feb 2010 LSE)
         • We have improved order:ack average by 30% & consistency by 90%
         • Multicast ack:tick has improved by 50%
   –   We continue to focus on improving consistency (see later slides for details)
• My Observations
   –   Since migration to Millennium IT, the exchange latency is an order of magnitude
       better than latency within client applications + transmission to/from our venues
         • Therefore for the majority of clients who are not co-located, further improvements to
           latency will have little impact
         • E.g. a 10% improvement to exchange latency = client 1% improvement
   –   Co-location customers whose algorithms interact exclusively with LSE/TQ markets
       have the most to gain from further reductions in latency
   –   Our latency is a tiny fraction (<0.0005%) of human reaction times (typically <200mS)

                                                                                                   6
Low Latency: Who really needs it?
    • Clients/Participants
 Category                Least Sensitive                            Most Sensitive

 Manual Trading          Brokers,                                   Arbitrage Traders,
                         “High Touch” Traders.                      Portfolio Traders.
 Automated Trading       Retail Service Providers (RSP),            Market Makers.
                         DMA Clients.
 Algorithmic engines &   Remote Algorithms e.g. VWAP, POV, IS       HFTs, Best Execution SORs,
 Smart Order Routers.    typically driven more by historical data   Co-lo Algorithms (e.g. arbitrage,
                         curves than real-time ticks.               momentum).


    • The Venue/Exchange
            –   Venues with the lowest latency Market Data are at the front of
                the queue when BBO updates occur
            –   Quality of dark book trades depends on ref price latency
            –   Builds strong technology reputation
            –   Generally good for Sales & marketing


                                                                                                        7
Low Latency: What’s the next best thing?

   • Consistency
      –   Fewer outliers, and outliers closer to the average rather than
          simply seeking ever lower average latency is the most important
          feature for any trading strategy whether manual or automated.
      –   If your system cannot compete on latency average, you can
          always make improvements by making it more consistent.

   • Resilience
      –   Closely tied to consistency is resilience, since an unresilient
          system will never be able to provide consistent results!!




                                                                            8
Low Latency: Technology

                       MIT/LSE/Turquoise        Customers
How did we get here?   •   C++                  • Fibre Optics
                       •   Linux                • Co-Location
                       •   Infiniband           • FPGA
                       •   Native + ITCH
                       •   1Gb Extranex
What next?             •   Rebalancing          •   More automation
                       •   More channels        •   Better Algorithms
                       •   More partitions      •   Faster switches
                       •   More threads         •   Faster firewalls
                       •   FPGA + processors    •   FPGA + processors
                       •   Tickerplant
                       •   Lower level API
                       •   Generation 8 …
                       •   R&D
                       •   Watch competitors!

                                                                        9
Performance & Capacity Management – KPIs

   #   KPI                                        Requirements / Target

   1   Total Daily Transactions                   Max (4 x average, 2 x peak)

   2   Total daily Trades                         Max (4 x average, 2 x peak)

   3   Order Latency                              Agreed levels per market

   4   Market Data Latency                        Agreed levels per market

   5   Reference Price Latency                    Agreed levels per market (TQ only)

   6   Transactions per second (1 sec peak)       Max (4 x average, 2 x peak)

   7   Transactions per second (10 sec average)   Max (4 x average, 2 x peak)

   8   Transactions per second (60 second         Max (4 x average, 2 x peak)
       average)
   9   System Availability                        100%




                                                                                       10
Performance & Capacity Management
  • Each system is technically tested with every software
    release to
      –   Prove KPI levels
      –   Reconfirm/prove behaviour and known breaking points
  • Preliminary tests take place on pre-production
  • At least one test cycle takes place on the actual
    production hardware on a weekend
  • Golden Rules
      1. A client message will always get a valid response (ack or nack)
      2. Any component that has been taken so far above the KPI level
         that it fails must do so gracefully, and the system must
         continue to obey golden rule 1.
   • Because we know the system/component limitations &
     bottlenecks we can manage growth as the markets
     evolve


                                                                           11
Performance & Capacity Management
Turquoise Lit Book Matching Engine Peak TPS last 12 months




                                                             12
To Summarise
• Low Latency
   –   What are we measuring?      (1) Order:Ack, (2) Ack:Tick, (3) Ref Prices.
   –   How Fast Is Fast Enough?    Millennium Average OK – Focus on Consistency.
   –   Who really needs it?        Predominantly Market Makers, HFTs, Co-Lo Algos.
   –   What’s the next best thing? Consistency and Resilience.
   –   The Technology
        • How did we get here?     C++/Linux/Infiniband/Native/ITCH
        • What next?               Balancing/threading/FPGA/Tickerplant,
                                   low level API/Generation 8 …
• Performance & Capacity Management
   –   The LSE/TQ KPIs             Trans,Trades,latencies,TPS (1/10/60), Availability.
   –   System behaviour under      1. Always respond, 2. Fail gracefully.
       extreme conditions




                                                                                         13
Questions?




             14

More Related Content

What's hot

Trading Platforms UI Clients Overview
Trading Platforms UI Clients OverviewTrading Platforms UI Clients Overview
Trading Platforms UI Clients OverviewVadim Pankin
 
Exactpro Test Tools EXTENT Feb 2011
Exactpro Test Tools EXTENT Feb 2011Exactpro Test Tools EXTENT Feb 2011
Exactpro Test Tools EXTENT Feb 2011Iosif Itkin
 
MIPI DevCon 2016: Effective Verification of Stacked and Layered Protocols
MIPI DevCon 2016: Effective Verification of Stacked and Layered ProtocolsMIPI DevCon 2016: Effective Verification of Stacked and Layered Protocols
MIPI DevCon 2016: Effective Verification of Stacked and Layered ProtocolsMIPI Alliance
 
How to create effective NOC in Poland
How to create effective NOC in PolandHow to create effective NOC in Poland
How to create effective NOC in PolandKamil Grabowski
 
FreeSWITCH Monitoring
FreeSWITCH MonitoringFreeSWITCH Monitoring
FreeSWITCH MonitoringMoises Silva
 
MIPI DevCon 2016: MIPI DisCo and ACPI - Streamlining MIPI Component Integration
MIPI DevCon 2016: MIPI DisCo and ACPI - Streamlining MIPI Component IntegrationMIPI DevCon 2016: MIPI DisCo and ACPI - Streamlining MIPI Component Integration
MIPI DevCon 2016: MIPI DisCo and ACPI - Streamlining MIPI Component IntegrationMIPI Alliance
 
Striving for ultimate Low Latency
Striving for ultimate Low LatencyStriving for ultimate Low Latency
Striving for ultimate Low LatencyMateusz Pusz
 
MIPI DevCon 2016: MIPI in Automotive
MIPI DevCon 2016: MIPI in AutomotiveMIPI DevCon 2016: MIPI in Automotive
MIPI DevCon 2016: MIPI in AutomotiveMIPI Alliance
 
Lahav Savir - Massively Scaleable Mobile Gateways
Lahav Savir - Massively Scaleable Mobile GatewaysLahav Savir - Massively Scaleable Mobile Gateways
Lahav Savir - Massively Scaleable Mobile GatewaysLahav Savir
 
2017 - LISA - LinkedIn's Distributed Firewall (DFW)
2017 - LISA - LinkedIn's Distributed Firewall (DFW)2017 - LISA - LinkedIn's Distributed Firewall (DFW)
2017 - LISA - LinkedIn's Distributed Firewall (DFW)Mike Svoboda
 
dat-TrafficManager-for-Vantage
dat-TrafficManager-for-Vantagedat-TrafficManager-for-Vantage
dat-TrafficManager-for-VantageScott Matics
 
Denovo SIP VoIP Termination SBC Session Boarder Controler @ denofolab.com
Denovo SIP VoIP Termination SBC Session Boarder Controler @ denofolab.comDenovo SIP VoIP Termination SBC Session Boarder Controler @ denofolab.com
Denovo SIP VoIP Termination SBC Session Boarder Controler @ denofolab.comAnne Kwong
 

What's hot (15)

Trading Platforms UI Clients Overview
Trading Platforms UI Clients OverviewTrading Platforms UI Clients Overview
Trading Platforms UI Clients Overview
 
Exactpro Test Tools EXTENT Feb 2011
Exactpro Test Tools EXTENT Feb 2011Exactpro Test Tools EXTENT Feb 2011
Exactpro Test Tools EXTENT Feb 2011
 
MIPI DevCon 2016: Effective Verification of Stacked and Layered Protocols
MIPI DevCon 2016: Effective Verification of Stacked and Layered ProtocolsMIPI DevCon 2016: Effective Verification of Stacked and Layered Protocols
MIPI DevCon 2016: Effective Verification of Stacked and Layered Protocols
 
How to create effective NOC in Poland
How to create effective NOC in PolandHow to create effective NOC in Poland
How to create effective NOC in Poland
 
FreeSWITCH Monitoring
FreeSWITCH MonitoringFreeSWITCH Monitoring
FreeSWITCH Monitoring
 
MIPI DevCon 2016: MIPI DisCo and ACPI - Streamlining MIPI Component Integration
MIPI DevCon 2016: MIPI DisCo and ACPI - Streamlining MIPI Component IntegrationMIPI DevCon 2016: MIPI DisCo and ACPI - Streamlining MIPI Component Integration
MIPI DevCon 2016: MIPI DisCo and ACPI - Streamlining MIPI Component Integration
 
Striving for ultimate Low Latency
Striving for ultimate Low LatencyStriving for ultimate Low Latency
Striving for ultimate Low Latency
 
MIPI DevCon 2016: MIPI in Automotive
MIPI DevCon 2016: MIPI in AutomotiveMIPI DevCon 2016: MIPI in Automotive
MIPI DevCon 2016: MIPI in Automotive
 
Lahav Savir - Massively Scaleable Mobile Gateways
Lahav Savir - Massively Scaleable Mobile GatewaysLahav Savir - Massively Scaleable Mobile Gateways
Lahav Savir - Massively Scaleable Mobile Gateways
 
MVTS PRO
MVTS PROMVTS PRO
MVTS PRO
 
2017 - LISA - LinkedIn's Distributed Firewall (DFW)
2017 - LISA - LinkedIn's Distributed Firewall (DFW)2017 - LISA - LinkedIn's Distributed Firewall (DFW)
2017 - LISA - LinkedIn's Distributed Firewall (DFW)
 
MVTS II
MVTS IIMVTS II
MVTS II
 
MVTS II Overview Presentation
MVTS II Overview PresentationMVTS II Overview Presentation
MVTS II Overview Presentation
 
dat-TrafficManager-for-Vantage
dat-TrafficManager-for-Vantagedat-TrafficManager-for-Vantage
dat-TrafficManager-for-Vantage
 
Denovo SIP VoIP Termination SBC Session Boarder Controler @ denofolab.com
Denovo SIP VoIP Termination SBC Session Boarder Controler @ denofolab.comDenovo SIP VoIP Termination SBC Session Boarder Controler @ denofolab.com
Denovo SIP VoIP Termination SBC Session Boarder Controler @ denofolab.com
 

Viewers also liked

Extent3 exante broker_for_algorithmic_trading_2012
Extent3 exante broker_for_algorithmic_trading_2012Extent3 exante broker_for_algorithmic_trading_2012
Extent3 exante broker_for_algorithmic_trading_2012extentconf Tsoy
 
Extent3 exactpro testing_of_hft_gui
Extent3 exactpro testing_of_hft_guiExtent3 exactpro testing_of_hft_gui
Extent3 exactpro testing_of_hft_guiextentconf Tsoy
 
Liquidity Fragmentation & SOR
Liquidity Fragmentation & SORLiquidity Fragmentation & SOR
Liquidity Fragmentation & SORIosif Itkin
 
QA Meet up in Saratov 25.07.15: Инструменты для тестирования и Лондонская фон...
QA Meet up in Saratov 25.07.15: Инструменты для тестирования и Лондонская фон...QA Meet up in Saratov 25.07.15: Инструменты для тестирования и Лондонская фон...
QA Meet up in Saratov 25.07.15: Инструменты для тестирования и Лондонская фон...Iosif Itkin
 
Extent 2013 Obninsk LSE - The Focus Beyond Low Latency
Extent 2013 Obninsk  LSE - The Focus Beyond Low LatencyExtent 2013 Obninsk  LSE - The Focus Beyond Low Latency
Extent 2013 Obninsk LSE - The Focus Beyond Low Latencyextentconf Tsoy
 
Extent 2013 Obninsk Managing Uncertain Data at Scale
Extent 2013 Obninsk Managing Uncertain Data at ScaleExtent 2013 Obninsk Managing Uncertain Data at Scale
Extent 2013 Obninsk Managing Uncertain Data at Scaleextentconf Tsoy
 
Extent 2013 Obninsk Test Tools for Trading Systems: Evolution Theory
Extent 2013 Obninsk Test Tools for Trading Systems: Evolution TheoryExtent 2013 Obninsk Test Tools for Trading Systems: Evolution Theory
Extent 2013 Obninsk Test Tools for Trading Systems: Evolution Theoryextentconf Tsoy
 
Extent 2013 Obninsk Trading Systems: Testing at the Confluence of FT & NFT
Extent 2013 Obninsk Trading Systems: Testing at the Confluence of FT & NFTExtent 2013 Obninsk Trading Systems: Testing at the Confluence of FT & NFT
Extent 2013 Obninsk Trading Systems: Testing at the Confluence of FT & NFTextentconf Tsoy
 
EXTENT-2015: Hyper-Fast Trading
EXTENT-2015: Hyper-Fast TradingEXTENT-2015: Hyper-Fast Trading
EXTENT-2015: Hyper-Fast TradingIosif Itkin
 
EXTENT-2015: A Test Harness for Algo Trading Systems
EXTENT-2015: A Test Harness for Algo Trading Systems EXTENT-2015: A Test Harness for Algo Trading Systems
EXTENT-2015: A Test Harness for Algo Trading Systems Iosif Itkin
 
EXTENT-2015: Blockchain New Frontiers
EXTENT-2015: Blockchain New FrontiersEXTENT-2015: Blockchain New Frontiers
EXTENT-2015: Blockchain New FrontiersIosif Itkin
 
EXTENT-2015: UnaVista Technology 
EXTENT-2015: UnaVista Technology EXTENT-2015: UnaVista Technology 
EXTENT-2015: UnaVista Technology Iosif Itkin
 
EXTENT-2015: Big Button 2.0
EXTENT-2015: Big Button 2.0EXTENT-2015: Big Button 2.0
EXTENT-2015: Big Button 2.0Iosif Itkin
 
EXTENT-2015: LSEG Technology Overview
EXTENT-2015: LSEG Technology Overview EXTENT-2015: LSEG Technology Overview
EXTENT-2015: LSEG Technology Overview Iosif Itkin
 
EXTENT-2015: Millennium Surveillance™ – Achieving Excellence
EXTENT-2015: Millennium Surveillance™ –  Achieving ExcellenceEXTENT-2015: Millennium Surveillance™ –  Achieving Excellence
EXTENT-2015: Millennium Surveillance™ – Achieving ExcellenceIosif Itkin
 
EXTENT-2016: Testing the Architecture
EXTENT-2016: Testing the ArchitectureEXTENT-2016: Testing the Architecture
EXTENT-2016: Testing the ArchitectureIosif Itkin
 
EXTENT-2015: Reconciliation Testing Aspects
EXTENT-2015: Reconciliation Testing AspectsEXTENT-2015: Reconciliation Testing Aspects
EXTENT-2015: Reconciliation Testing AspectsIosif Itkin
 
Extent3 prognoz practical_approach_lppl_model_2012
Extent3 prognoz practical_approach_lppl_model_2012Extent3 prognoz practical_approach_lppl_model_2012
Extent3 prognoz practical_approach_lppl_model_2012extentconf Tsoy
 
EXTENT-2016: Trading Technology Trends and Innovation
EXTENT-2016: Trading Technology Trends and InnovationEXTENT-2016: Trading Technology Trends and Innovation
EXTENT-2016: Trading Technology Trends and InnovationIosif Itkin
 
EXTENT-2016: Industry Practices of Advanced Program Analysis
EXTENT-2016: Industry Practices of Advanced Program AnalysisEXTENT-2016: Industry Practices of Advanced Program Analysis
EXTENT-2016: Industry Practices of Advanced Program AnalysisIosif Itkin
 

Viewers also liked (20)

Extent3 exante broker_for_algorithmic_trading_2012
Extent3 exante broker_for_algorithmic_trading_2012Extent3 exante broker_for_algorithmic_trading_2012
Extent3 exante broker_for_algorithmic_trading_2012
 
Extent3 exactpro testing_of_hft_gui
Extent3 exactpro testing_of_hft_guiExtent3 exactpro testing_of_hft_gui
Extent3 exactpro testing_of_hft_gui
 
Liquidity Fragmentation & SOR
Liquidity Fragmentation & SORLiquidity Fragmentation & SOR
Liquidity Fragmentation & SOR
 
QA Meet up in Saratov 25.07.15: Инструменты для тестирования и Лондонская фон...
QA Meet up in Saratov 25.07.15: Инструменты для тестирования и Лондонская фон...QA Meet up in Saratov 25.07.15: Инструменты для тестирования и Лондонская фон...
QA Meet up in Saratov 25.07.15: Инструменты для тестирования и Лондонская фон...
 
Extent 2013 Obninsk LSE - The Focus Beyond Low Latency
Extent 2013 Obninsk  LSE - The Focus Beyond Low LatencyExtent 2013 Obninsk  LSE - The Focus Beyond Low Latency
Extent 2013 Obninsk LSE - The Focus Beyond Low Latency
 
Extent 2013 Obninsk Managing Uncertain Data at Scale
Extent 2013 Obninsk Managing Uncertain Data at ScaleExtent 2013 Obninsk Managing Uncertain Data at Scale
Extent 2013 Obninsk Managing Uncertain Data at Scale
 
Extent 2013 Obninsk Test Tools for Trading Systems: Evolution Theory
Extent 2013 Obninsk Test Tools for Trading Systems: Evolution TheoryExtent 2013 Obninsk Test Tools for Trading Systems: Evolution Theory
Extent 2013 Obninsk Test Tools for Trading Systems: Evolution Theory
 
Extent 2013 Obninsk Trading Systems: Testing at the Confluence of FT & NFT
Extent 2013 Obninsk Trading Systems: Testing at the Confluence of FT & NFTExtent 2013 Obninsk Trading Systems: Testing at the Confluence of FT & NFT
Extent 2013 Obninsk Trading Systems: Testing at the Confluence of FT & NFT
 
EXTENT-2015: Hyper-Fast Trading
EXTENT-2015: Hyper-Fast TradingEXTENT-2015: Hyper-Fast Trading
EXTENT-2015: Hyper-Fast Trading
 
EXTENT-2015: A Test Harness for Algo Trading Systems
EXTENT-2015: A Test Harness for Algo Trading Systems EXTENT-2015: A Test Harness for Algo Trading Systems
EXTENT-2015: A Test Harness for Algo Trading Systems
 
EXTENT-2015: Blockchain New Frontiers
EXTENT-2015: Blockchain New FrontiersEXTENT-2015: Blockchain New Frontiers
EXTENT-2015: Blockchain New Frontiers
 
EXTENT-2015: UnaVista Technology 
EXTENT-2015: UnaVista Technology EXTENT-2015: UnaVista Technology 
EXTENT-2015: UnaVista Technology 
 
EXTENT-2015: Big Button 2.0
EXTENT-2015: Big Button 2.0EXTENT-2015: Big Button 2.0
EXTENT-2015: Big Button 2.0
 
EXTENT-2015: LSEG Technology Overview
EXTENT-2015: LSEG Technology Overview EXTENT-2015: LSEG Technology Overview
EXTENT-2015: LSEG Technology Overview
 
EXTENT-2015: Millennium Surveillance™ – Achieving Excellence
EXTENT-2015: Millennium Surveillance™ –  Achieving ExcellenceEXTENT-2015: Millennium Surveillance™ –  Achieving Excellence
EXTENT-2015: Millennium Surveillance™ – Achieving Excellence
 
EXTENT-2016: Testing the Architecture
EXTENT-2016: Testing the ArchitectureEXTENT-2016: Testing the Architecture
EXTENT-2016: Testing the Architecture
 
EXTENT-2015: Reconciliation Testing Aspects
EXTENT-2015: Reconciliation Testing AspectsEXTENT-2015: Reconciliation Testing Aspects
EXTENT-2015: Reconciliation Testing Aspects
 
Extent3 prognoz practical_approach_lppl_model_2012
Extent3 prognoz practical_approach_lppl_model_2012Extent3 prognoz practical_approach_lppl_model_2012
Extent3 prognoz practical_approach_lppl_model_2012
 
EXTENT-2016: Trading Technology Trends and Innovation
EXTENT-2016: Trading Technology Trends and InnovationEXTENT-2016: Trading Technology Trends and Innovation
EXTENT-2016: Trading Technology Trends and Innovation
 
EXTENT-2016: Industry Practices of Advanced Program Analysis
EXTENT-2016: Industry Practices of Advanced Program AnalysisEXTENT-2016: Industry Practices of Advanced Program Analysis
EXTENT-2016: Industry Practices of Advanced Program Analysis
 

Similar to Extent3 turquoise equity_trading_2012

Protecting Your API with Redis by Jane Paek - Redis Day Seattle 2020
Protecting Your API with Redis by Jane Paek - Redis Day Seattle 2020Protecting Your API with Redis by Jane Paek - Redis Day Seattle 2020
Protecting Your API with Redis by Jane Paek - Redis Day Seattle 2020Redis Labs
 
Mission critical computing by intel
Mission critical computing by intelMission critical computing by intel
Mission critical computing by intelHP ESSN Philippines
 
Scaling habits of ASP.NET
Scaling habits of ASP.NETScaling habits of ASP.NET
Scaling habits of ASP.NETDavid Giard
 
Tsinghua University: Two Exemplary Applications in China
Tsinghua University: Two Exemplary Applications in ChinaTsinghua University: Two Exemplary Applications in China
Tsinghua University: Two Exemplary Applications in ChinaDataStax Academy
 
You name it, we analyze it
You name it, we analyze itYou name it, we analyze it
You name it, we analyze itJim Gilsinn
 
(DVO205) Monitoring Evolution: Flying Blind to Flying by Instrument
(DVO205) Monitoring Evolution: Flying Blind to Flying by Instrument(DVO205) Monitoring Evolution: Flying Blind to Flying by Instrument
(DVO205) Monitoring Evolution: Flying Blind to Flying by InstrumentAmazon Web Services
 
Dev Lakhani, Data Scientist at Batch Insights "Real Time Big Data Applicatio...
Dev Lakhani, Data Scientist at Batch Insights  "Real Time Big Data Applicatio...Dev Lakhani, Data Scientist at Batch Insights  "Real Time Big Data Applicatio...
Dev Lakhani, Data Scientist at Batch Insights "Real Time Big Data Applicatio...Dataconomy Media
 
Test scenario simulator
Test scenario simulatorTest scenario simulator
Test scenario simulatortahmed
 
Test scenario simulator
Test scenario simulatorTest scenario simulator
Test scenario simulatorguest4ebcd7b
 
Metrics driven development with dedicated Observability Team
Metrics driven development with dedicated Observability TeamMetrics driven development with dedicated Observability Team
Metrics driven development with dedicated Observability TeamLINE Corporation
 
A Practical Guide to Selecting a Stream Processing Technology
A Practical Guide to Selecting a Stream Processing Technology A Practical Guide to Selecting a Stream Processing Technology
A Practical Guide to Selecting a Stream Processing Technology confluent
 
Non-Functional Requirements
Non-Functional RequirementsNon-Functional Requirements
Non-Functional RequirementsDavid Simons
 
DevDay: Corda Enterprise: Journey to 1000 TPS per node, Rick Parker
DevDay: Corda Enterprise: Journey to 1000 TPS per node, Rick ParkerDevDay: Corda Enterprise: Journey to 1000 TPS per node, Rick Parker
DevDay: Corda Enterprise: Journey to 1000 TPS per node, Rick ParkerR3
 
Time Critical Networks
Time Critical NetworksTime Critical Networks
Time Critical NetworksLars Bröhne
 

Similar to Extent3 turquoise equity_trading_2012 (20)

Open Daylight Forum India 2015
Open Daylight Forum India 2015Open Daylight Forum India 2015
Open Daylight Forum India 2015
 
Kafka at scale facebook israel
Kafka at scale   facebook israelKafka at scale   facebook israel
Kafka at scale facebook israel
 
Protecting Your API with Redis by Jane Paek - Redis Day Seattle 2020
Protecting Your API with Redis by Jane Paek - Redis Day Seattle 2020Protecting Your API with Redis by Jane Paek - Redis Day Seattle 2020
Protecting Your API with Redis by Jane Paek - Redis Day Seattle 2020
 
Mission critical computing by intel
Mission critical computing by intelMission critical computing by intel
Mission critical computing by intel
 
Scaling habits of ASP.NET
Scaling habits of ASP.NETScaling habits of ASP.NET
Scaling habits of ASP.NET
 
Chapter02
Chapter02Chapter02
Chapter02
 
Tsinghua University: Two Exemplary Applications in China
Tsinghua University: Two Exemplary Applications in ChinaTsinghua University: Two Exemplary Applications in China
Tsinghua University: Two Exemplary Applications in China
 
Play With Streams
Play With StreamsPlay With Streams
Play With Streams
 
You name it, we analyze it
You name it, we analyze itYou name it, we analyze it
You name it, we analyze it
 
Rate limits and all about
Rate limits and all aboutRate limits and all about
Rate limits and all about
 
(DVO205) Monitoring Evolution: Flying Blind to Flying by Instrument
(DVO205) Monitoring Evolution: Flying Blind to Flying by Instrument(DVO205) Monitoring Evolution: Flying Blind to Flying by Instrument
(DVO205) Monitoring Evolution: Flying Blind to Flying by Instrument
 
Dev Lakhani, Data Scientist at Batch Insights "Real Time Big Data Applicatio...
Dev Lakhani, Data Scientist at Batch Insights  "Real Time Big Data Applicatio...Dev Lakhani, Data Scientist at Batch Insights  "Real Time Big Data Applicatio...
Dev Lakhani, Data Scientist at Batch Insights "Real Time Big Data Applicatio...
 
Test scenario simulator
Test scenario simulatorTest scenario simulator
Test scenario simulator
 
Test scenario simulator
Test scenario simulatorTest scenario simulator
Test scenario simulator
 
Metrics driven development with dedicated Observability Team
Metrics driven development with dedicated Observability TeamMetrics driven development with dedicated Observability Team
Metrics driven development with dedicated Observability Team
 
A Practical Guide to Selecting a Stream Processing Technology
A Practical Guide to Selecting a Stream Processing Technology A Practical Guide to Selecting a Stream Processing Technology
A Practical Guide to Selecting a Stream Processing Technology
 
Non-Functional Requirements
Non-Functional RequirementsNon-Functional Requirements
Non-Functional Requirements
 
Algorithmic Trading
Algorithmic TradingAlgorithmic Trading
Algorithmic Trading
 
DevDay: Corda Enterprise: Journey to 1000 TPS per node, Rick Parker
DevDay: Corda Enterprise: Journey to 1000 TPS per node, Rick ParkerDevDay: Corda Enterprise: Journey to 1000 TPS per node, Rick Parker
DevDay: Corda Enterprise: Journey to 1000 TPS per node, Rick Parker
 
Time Critical Networks
Time Critical NetworksTime Critical Networks
Time Critical Networks
 

More from extentconf Tsoy

Extent 2013 Obninsk How a Great QA Team Can Make a Disproportionate Contribut...
Extent 2013 Obninsk How a Great QA Team Can Make a Disproportionate Contribut...Extent 2013 Obninsk How a Great QA Team Can Make a Disproportionate Contribut...
Extent 2013 Obninsk How a Great QA Team Can Make a Disproportionate Contribut...extentconf Tsoy
 
Extent 2013 Obninsk Cross-Asset Portfolio Margin Risk Calculation for HFT
Extent 2013 Obninsk Cross-Asset Portfolio Margin Risk Calculation for HFTExtent 2013 Obninsk Cross-Asset Portfolio Margin Risk Calculation for HFT
Extent 2013 Obninsk Cross-Asset Portfolio Margin Risk Calculation for HFTextentconf Tsoy
 
Extent april2012-kostroma social-networks-socialmedia-trading
Extent april2012-kostroma social-networks-socialmedia-tradingExtent april2012-kostroma social-networks-socialmedia-trading
Extent april2012-kostroma social-networks-socialmedia-tradingextentconf Tsoy
 
Extent3 exactpro the_future_of_risk_controls
Extent3 exactpro the_future_of_risk_controlsExtent3 exactpro the_future_of_risk_controls
Extent3 exactpro the_future_of_risk_controlsextentconf Tsoy
 
Extent3 exactpro four_houses_test_tools_2012 (1)
Extent3 exactpro four_houses_test_tools_2012 (1)Extent3 exactpro four_houses_test_tools_2012 (1)
Extent3 exactpro four_houses_test_tools_2012 (1)extentconf Tsoy
 
Extent3 witology prediction_markets_2012
Extent3 witology prediction_markets_2012Extent3 witology prediction_markets_2012
Extent3 witology prediction_markets_2012extentconf Tsoy
 
Extent3 exactpro the_next_step_in_reconciliation_testing
Extent3 exactpro the_next_step_in_reconciliation_testingExtent3 exactpro the_next_step_in_reconciliation_testing
Extent3 exactpro the_next_step_in_reconciliation_testingextentconf Tsoy
 
Verification of Financial Models
Verification of Financial ModelsVerification of Financial Models
Verification of Financial Modelsextentconf Tsoy
 
The Simple Matter of Project Management
The Simple Matter of Project ManagementThe Simple Matter of Project Management
The Simple Matter of Project Managementextentconf Tsoy
 
Exchange Simulators for SOR / Algo Testing: Advantages vs. Shortcomings
Exchange Simulators for SOR / Algo Testing: Advantages vs. ShortcomingsExchange Simulators for SOR / Algo Testing: Advantages vs. Shortcomings
Exchange Simulators for SOR / Algo Testing: Advantages vs. Shortcomingsextentconf Tsoy
 
Behavior Driven Development Pros and Cons
Behavior Driven Development Pros and ConsBehavior Driven Development Pros and Cons
Behavior Driven Development Pros and Consextentconf Tsoy
 
Virtualization Technology for Test Automation
Virtualization Technology for Test AutomationVirtualization Technology for Test Automation
Virtualization Technology for Test Automationextentconf Tsoy
 
Cost of Quality How to Save Money
Cost of Quality How to Save MoneyCost of Quality How to Save Money
Cost of Quality How to Save Moneyextentconf Tsoy
 
Exactpro Test Tools EXTENT Feb 2011
Exactpro Test Tools EXTENT Feb 2011Exactpro Test Tools EXTENT Feb 2011
Exactpro Test Tools EXTENT Feb 2011extentconf Tsoy
 
Technical Testing Introduction
Technical Testing IntroductionTechnical Testing Introduction
Technical Testing Introductionextentconf Tsoy
 
Financial Instruments EXTENT February 2011
Financial Instruments EXTENT February 2011Financial Instruments EXTENT February 2011
Financial Instruments EXTENT February 2011extentconf Tsoy
 
Liquidity Fragmentation & SOR
Liquidity Fragmentation & SORLiquidity Fragmentation & SOR
Liquidity Fragmentation & SORextentconf Tsoy
 

More from extentconf Tsoy (17)

Extent 2013 Obninsk How a Great QA Team Can Make a Disproportionate Contribut...
Extent 2013 Obninsk How a Great QA Team Can Make a Disproportionate Contribut...Extent 2013 Obninsk How a Great QA Team Can Make a Disproportionate Contribut...
Extent 2013 Obninsk How a Great QA Team Can Make a Disproportionate Contribut...
 
Extent 2013 Obninsk Cross-Asset Portfolio Margin Risk Calculation for HFT
Extent 2013 Obninsk Cross-Asset Portfolio Margin Risk Calculation for HFTExtent 2013 Obninsk Cross-Asset Portfolio Margin Risk Calculation for HFT
Extent 2013 Obninsk Cross-Asset Portfolio Margin Risk Calculation for HFT
 
Extent april2012-kostroma social-networks-socialmedia-trading
Extent april2012-kostroma social-networks-socialmedia-tradingExtent april2012-kostroma social-networks-socialmedia-trading
Extent april2012-kostroma social-networks-socialmedia-trading
 
Extent3 exactpro the_future_of_risk_controls
Extent3 exactpro the_future_of_risk_controlsExtent3 exactpro the_future_of_risk_controls
Extent3 exactpro the_future_of_risk_controls
 
Extent3 exactpro four_houses_test_tools_2012 (1)
Extent3 exactpro four_houses_test_tools_2012 (1)Extent3 exactpro four_houses_test_tools_2012 (1)
Extent3 exactpro four_houses_test_tools_2012 (1)
 
Extent3 witology prediction_markets_2012
Extent3 witology prediction_markets_2012Extent3 witology prediction_markets_2012
Extent3 witology prediction_markets_2012
 
Extent3 exactpro the_next_step_in_reconciliation_testing
Extent3 exactpro the_next_step_in_reconciliation_testingExtent3 exactpro the_next_step_in_reconciliation_testing
Extent3 exactpro the_next_step_in_reconciliation_testing
 
Verification of Financial Models
Verification of Financial ModelsVerification of Financial Models
Verification of Financial Models
 
The Simple Matter of Project Management
The Simple Matter of Project ManagementThe Simple Matter of Project Management
The Simple Matter of Project Management
 
Exchange Simulators for SOR / Algo Testing: Advantages vs. Shortcomings
Exchange Simulators for SOR / Algo Testing: Advantages vs. ShortcomingsExchange Simulators for SOR / Algo Testing: Advantages vs. Shortcomings
Exchange Simulators for SOR / Algo Testing: Advantages vs. Shortcomings
 
Behavior Driven Development Pros and Cons
Behavior Driven Development Pros and ConsBehavior Driven Development Pros and Cons
Behavior Driven Development Pros and Cons
 
Virtualization Technology for Test Automation
Virtualization Technology for Test AutomationVirtualization Technology for Test Automation
Virtualization Technology for Test Automation
 
Cost of Quality How to Save Money
Cost of Quality How to Save MoneyCost of Quality How to Save Money
Cost of Quality How to Save Money
 
Exactpro Test Tools EXTENT Feb 2011
Exactpro Test Tools EXTENT Feb 2011Exactpro Test Tools EXTENT Feb 2011
Exactpro Test Tools EXTENT Feb 2011
 
Technical Testing Introduction
Technical Testing IntroductionTechnical Testing Introduction
Technical Testing Introduction
 
Financial Instruments EXTENT February 2011
Financial Instruments EXTENT February 2011Financial Instruments EXTENT February 2011
Financial Instruments EXTENT February 2011
 
Liquidity Fragmentation & SOR
Liquidity Fragmentation & SORLiquidity Fragmentation & SOR
Liquidity Fragmentation & SOR
 

Recently uploaded

Tampa BSides - Chef's Tour of Microsoft Security Adoption Framework (SAF)
Tampa BSides - Chef's Tour of Microsoft Security Adoption Framework (SAF)Tampa BSides - Chef's Tour of Microsoft Security Adoption Framework (SAF)
Tampa BSides - Chef's Tour of Microsoft Security Adoption Framework (SAF)Mark Simos
 
Bun (KitWorks Team Study 노별마루 발표 2024.4.22)
Bun (KitWorks Team Study 노별마루 발표 2024.4.22)Bun (KitWorks Team Study 노별마루 발표 2024.4.22)
Bun (KitWorks Team Study 노별마루 발표 2024.4.22)Wonjun Hwang
 
Advanced Test Driven-Development @ php[tek] 2024
Advanced Test Driven-Development @ php[tek] 2024Advanced Test Driven-Development @ php[tek] 2024
Advanced Test Driven-Development @ php[tek] 2024Scott Keck-Warren
 
"Debugging python applications inside k8s environment", Andrii Soldatenko
"Debugging python applications inside k8s environment", Andrii Soldatenko"Debugging python applications inside k8s environment", Andrii Soldatenko
"Debugging python applications inside k8s environment", Andrii SoldatenkoFwdays
 
Leverage Zilliz Serverless - Up to 50X Saving for Your Vector Storage Cost
Leverage Zilliz Serverless - Up to 50X Saving for Your Vector Storage CostLeverage Zilliz Serverless - Up to 50X Saving for Your Vector Storage Cost
Leverage Zilliz Serverless - Up to 50X Saving for Your Vector Storage CostZilliz
 
SIP trunking in Janus @ Kamailio World 2024
SIP trunking in Janus @ Kamailio World 2024SIP trunking in Janus @ Kamailio World 2024
SIP trunking in Janus @ Kamailio World 2024Lorenzo Miniero
 
Scanning the Internet for External Cloud Exposures via SSL Certs
Scanning the Internet for External Cloud Exposures via SSL CertsScanning the Internet for External Cloud Exposures via SSL Certs
Scanning the Internet for External Cloud Exposures via SSL CertsRizwan Syed
 
New from BookNet Canada for 2024: BNC CataList - Tech Forum 2024
New from BookNet Canada for 2024: BNC CataList - Tech Forum 2024New from BookNet Canada for 2024: BNC CataList - Tech Forum 2024
New from BookNet Canada for 2024: BNC CataList - Tech Forum 2024BookNet Canada
 
"Subclassing and Composition – A Pythonic Tour of Trade-Offs", Hynek Schlawack
"Subclassing and Composition – A Pythonic Tour of Trade-Offs", Hynek Schlawack"Subclassing and Composition – A Pythonic Tour of Trade-Offs", Hynek Schlawack
"Subclassing and Composition – A Pythonic Tour of Trade-Offs", Hynek SchlawackFwdays
 
Vector Databases 101 - An introduction to the world of Vector Databases
Vector Databases 101 - An introduction to the world of Vector DatabasesVector Databases 101 - An introduction to the world of Vector Databases
Vector Databases 101 - An introduction to the world of Vector DatabasesZilliz
 
"LLMs for Python Engineers: Advanced Data Analysis and Semantic Kernel",Oleks...
"LLMs for Python Engineers: Advanced Data Analysis and Semantic Kernel",Oleks..."LLMs for Python Engineers: Advanced Data Analysis and Semantic Kernel",Oleks...
"LLMs for Python Engineers: Advanced Data Analysis and Semantic Kernel",Oleks...Fwdays
 
Integration and Automation in Practice: CI/CD in Mule Integration and Automat...
Integration and Automation in Practice: CI/CD in Mule Integration and Automat...Integration and Automation in Practice: CI/CD in Mule Integration and Automat...
Integration and Automation in Practice: CI/CD in Mule Integration and Automat...Patryk Bandurski
 
Unleash Your Potential - Namagunga Girls Coding Club
Unleash Your Potential - Namagunga Girls Coding ClubUnleash Your Potential - Namagunga Girls Coding Club
Unleash Your Potential - Namagunga Girls Coding ClubKalema Edgar
 
Dev Dives: Streamline document processing with UiPath Studio Web
Dev Dives: Streamline document processing with UiPath Studio WebDev Dives: Streamline document processing with UiPath Studio Web
Dev Dives: Streamline document processing with UiPath Studio WebUiPathCommunity
 
Vertex AI Gemini Prompt Engineering Tips
Vertex AI Gemini Prompt Engineering TipsVertex AI Gemini Prompt Engineering Tips
Vertex AI Gemini Prompt Engineering TipsMiki Katsuragi
 
Human Factors of XR: Using Human Factors to Design XR Systems
Human Factors of XR: Using Human Factors to Design XR SystemsHuman Factors of XR: Using Human Factors to Design XR Systems
Human Factors of XR: Using Human Factors to Design XR SystemsMark Billinghurst
 
Install Stable Diffusion in windows machine
Install Stable Diffusion in windows machineInstall Stable Diffusion in windows machine
Install Stable Diffusion in windows machinePadma Pradeep
 
SAP Build Work Zone - Overview L2-L3.pptx
SAP Build Work Zone - Overview L2-L3.pptxSAP Build Work Zone - Overview L2-L3.pptx
SAP Build Work Zone - Overview L2-L3.pptxNavinnSomaal
 
Are Multi-Cloud and Serverless Good or Bad?
Are Multi-Cloud and Serverless Good or Bad?Are Multi-Cloud and Serverless Good or Bad?
Are Multi-Cloud and Serverless Good or Bad?Mattias Andersson
 

Recently uploaded (20)

Tampa BSides - Chef's Tour of Microsoft Security Adoption Framework (SAF)
Tampa BSides - Chef's Tour of Microsoft Security Adoption Framework (SAF)Tampa BSides - Chef's Tour of Microsoft Security Adoption Framework (SAF)
Tampa BSides - Chef's Tour of Microsoft Security Adoption Framework (SAF)
 
Bun (KitWorks Team Study 노별마루 발표 2024.4.22)
Bun (KitWorks Team Study 노별마루 발표 2024.4.22)Bun (KitWorks Team Study 노별마루 발표 2024.4.22)
Bun (KitWorks Team Study 노별마루 발표 2024.4.22)
 
Advanced Test Driven-Development @ php[tek] 2024
Advanced Test Driven-Development @ php[tek] 2024Advanced Test Driven-Development @ php[tek] 2024
Advanced Test Driven-Development @ php[tek] 2024
 
"Debugging python applications inside k8s environment", Andrii Soldatenko
"Debugging python applications inside k8s environment", Andrii Soldatenko"Debugging python applications inside k8s environment", Andrii Soldatenko
"Debugging python applications inside k8s environment", Andrii Soldatenko
 
Leverage Zilliz Serverless - Up to 50X Saving for Your Vector Storage Cost
Leverage Zilliz Serverless - Up to 50X Saving for Your Vector Storage CostLeverage Zilliz Serverless - Up to 50X Saving for Your Vector Storage Cost
Leverage Zilliz Serverless - Up to 50X Saving for Your Vector Storage Cost
 
SIP trunking in Janus @ Kamailio World 2024
SIP trunking in Janus @ Kamailio World 2024SIP trunking in Janus @ Kamailio World 2024
SIP trunking in Janus @ Kamailio World 2024
 
Scanning the Internet for External Cloud Exposures via SSL Certs
Scanning the Internet for External Cloud Exposures via SSL CertsScanning the Internet for External Cloud Exposures via SSL Certs
Scanning the Internet for External Cloud Exposures via SSL Certs
 
New from BookNet Canada for 2024: BNC CataList - Tech Forum 2024
New from BookNet Canada for 2024: BNC CataList - Tech Forum 2024New from BookNet Canada for 2024: BNC CataList - Tech Forum 2024
New from BookNet Canada for 2024: BNC CataList - Tech Forum 2024
 
"Subclassing and Composition – A Pythonic Tour of Trade-Offs", Hynek Schlawack
"Subclassing and Composition – A Pythonic Tour of Trade-Offs", Hynek Schlawack"Subclassing and Composition – A Pythonic Tour of Trade-Offs", Hynek Schlawack
"Subclassing and Composition – A Pythonic Tour of Trade-Offs", Hynek Schlawack
 
Vector Databases 101 - An introduction to the world of Vector Databases
Vector Databases 101 - An introduction to the world of Vector DatabasesVector Databases 101 - An introduction to the world of Vector Databases
Vector Databases 101 - An introduction to the world of Vector Databases
 
"LLMs for Python Engineers: Advanced Data Analysis and Semantic Kernel",Oleks...
"LLMs for Python Engineers: Advanced Data Analysis and Semantic Kernel",Oleks..."LLMs for Python Engineers: Advanced Data Analysis and Semantic Kernel",Oleks...
"LLMs for Python Engineers: Advanced Data Analysis and Semantic Kernel",Oleks...
 
Integration and Automation in Practice: CI/CD in Mule Integration and Automat...
Integration and Automation in Practice: CI/CD in Mule Integration and Automat...Integration and Automation in Practice: CI/CD in Mule Integration and Automat...
Integration and Automation in Practice: CI/CD in Mule Integration and Automat...
 
Unleash Your Potential - Namagunga Girls Coding Club
Unleash Your Potential - Namagunga Girls Coding ClubUnleash Your Potential - Namagunga Girls Coding Club
Unleash Your Potential - Namagunga Girls Coding Club
 
Dev Dives: Streamline document processing with UiPath Studio Web
Dev Dives: Streamline document processing with UiPath Studio WebDev Dives: Streamline document processing with UiPath Studio Web
Dev Dives: Streamline document processing with UiPath Studio Web
 
Vertex AI Gemini Prompt Engineering Tips
Vertex AI Gemini Prompt Engineering TipsVertex AI Gemini Prompt Engineering Tips
Vertex AI Gemini Prompt Engineering Tips
 
Human Factors of XR: Using Human Factors to Design XR Systems
Human Factors of XR: Using Human Factors to Design XR SystemsHuman Factors of XR: Using Human Factors to Design XR Systems
Human Factors of XR: Using Human Factors to Design XR Systems
 
E-Vehicle_Hacking_by_Parul Sharma_null_owasp.pptx
E-Vehicle_Hacking_by_Parul Sharma_null_owasp.pptxE-Vehicle_Hacking_by_Parul Sharma_null_owasp.pptx
E-Vehicle_Hacking_by_Parul Sharma_null_owasp.pptx
 
Install Stable Diffusion in windows machine
Install Stable Diffusion in windows machineInstall Stable Diffusion in windows machine
Install Stable Diffusion in windows machine
 
SAP Build Work Zone - Overview L2-L3.pptx
SAP Build Work Zone - Overview L2-L3.pptxSAP Build Work Zone - Overview L2-L3.pptx
SAP Build Work Zone - Overview L2-L3.pptx
 
Are Multi-Cloud and Serverless Good or Bad?
Are Multi-Cloud and Serverless Good or Bad?Are Multi-Cloud and Serverless Good or Bad?
Are Multi-Cloud and Serverless Good or Bad?
 

Extent3 turquoise equity_trading_2012

  • 1. Equity Trading: Performance, Latency & Throughput Prepared for the 2012 Exactpro ExTent Conference, Kostroma, Russia. 11° April 2012 Author: Phil Penhaligan
  • 2. Agenda • Low Latency – What are we measuring? – How Fast Is Fast Enough? – Who really needs it? – What’s the next best thing? – The Technology • How did we get here? • What next? • Performance & Capacity Management – The LSE/TQ KPIs – System behaviour under extreme conditions 2
  • 3. Low Latency: What are we measuring? • 3 distinct attributes – Measured from just outside LSE/TQ firewalls – Excludes transmission time to/from external client/venue – Excludes latency of client/venue’s own system(s) – Excludes network packet retransmissions due to slow consumers 1. Order latency • Time from Order receipt to order acknowledgment • Commonly referred to as Order:Ack 2. Market Data latency • Time from Order acknowledgement to public broadcast Commonly referred to as Ack:Tick 3. Reference price latency (Turquoise Dark Book only) • Time from Tick receipt to matching engine usage 3
  • 4. Low Latency: What are we measuring? • 2 key metrics – Average • Mathematical mean – Consistency • Max 99.9 percentile – Discard the worst 0.1% of orders and then take the maximum • Production Systems – Continuous real-time capture – Historical data stored in database for statistical analysis and reporting • QA Systems – New software releases evaluated during technical tests – To confirm/verify expected latency improvements – To detect unexpected latency regressions 4
  • 5. Low Latency: Example – Turquoise Order:Ack Latency Distribution Average and Consistency Metric uS Average 100% 108 Average 99.9% 103 Max 99.9% 381 Max 99 191 5
  • 6. Low Latency: How Fast Is Fast Enough? • Customer feedback – LSE/TQ averages are already fast enough – better Consistency is most important! • Our Focus – Since the initial deployment of Millennium (Oct 2009 Turquoise and Feb 2010 LSE) • We have improved order:ack average by 30% & consistency by 90% • Multicast ack:tick has improved by 50% – We continue to focus on improving consistency (see later slides for details) • My Observations – Since migration to Millennium IT, the exchange latency is an order of magnitude better than latency within client applications + transmission to/from our venues • Therefore for the majority of clients who are not co-located, further improvements to latency will have little impact • E.g. a 10% improvement to exchange latency = client 1% improvement – Co-location customers whose algorithms interact exclusively with LSE/TQ markets have the most to gain from further reductions in latency – Our latency is a tiny fraction (<0.0005%) of human reaction times (typically <200mS) 6
  • 7. Low Latency: Who really needs it? • Clients/Participants Category Least Sensitive Most Sensitive Manual Trading Brokers, Arbitrage Traders, “High Touch” Traders. Portfolio Traders. Automated Trading Retail Service Providers (RSP), Market Makers. DMA Clients. Algorithmic engines & Remote Algorithms e.g. VWAP, POV, IS HFTs, Best Execution SORs, Smart Order Routers. typically driven more by historical data Co-lo Algorithms (e.g. arbitrage, curves than real-time ticks. momentum). • The Venue/Exchange – Venues with the lowest latency Market Data are at the front of the queue when BBO updates occur – Quality of dark book trades depends on ref price latency – Builds strong technology reputation – Generally good for Sales & marketing 7
  • 8. Low Latency: What’s the next best thing? • Consistency – Fewer outliers, and outliers closer to the average rather than simply seeking ever lower average latency is the most important feature for any trading strategy whether manual or automated. – If your system cannot compete on latency average, you can always make improvements by making it more consistent. • Resilience – Closely tied to consistency is resilience, since an unresilient system will never be able to provide consistent results!! 8
  • 9. Low Latency: Technology MIT/LSE/Turquoise Customers How did we get here? • C++ • Fibre Optics • Linux • Co-Location • Infiniband • FPGA • Native + ITCH • 1Gb Extranex What next? • Rebalancing • More automation • More channels • Better Algorithms • More partitions • Faster switches • More threads • Faster firewalls • FPGA + processors • FPGA + processors • Tickerplant • Lower level API • Generation 8 … • R&D • Watch competitors! 9
  • 10. Performance & Capacity Management – KPIs # KPI Requirements / Target 1 Total Daily Transactions Max (4 x average, 2 x peak) 2 Total daily Trades Max (4 x average, 2 x peak) 3 Order Latency Agreed levels per market 4 Market Data Latency Agreed levels per market 5 Reference Price Latency Agreed levels per market (TQ only) 6 Transactions per second (1 sec peak) Max (4 x average, 2 x peak) 7 Transactions per second (10 sec average) Max (4 x average, 2 x peak) 8 Transactions per second (60 second Max (4 x average, 2 x peak) average) 9 System Availability 100% 10
  • 11. Performance & Capacity Management • Each system is technically tested with every software release to – Prove KPI levels – Reconfirm/prove behaviour and known breaking points • Preliminary tests take place on pre-production • At least one test cycle takes place on the actual production hardware on a weekend • Golden Rules 1. A client message will always get a valid response (ack or nack) 2. Any component that has been taken so far above the KPI level that it fails must do so gracefully, and the system must continue to obey golden rule 1. • Because we know the system/component limitations & bottlenecks we can manage growth as the markets evolve 11
  • 12. Performance & Capacity Management Turquoise Lit Book Matching Engine Peak TPS last 12 months 12
  • 13. To Summarise • Low Latency – What are we measuring? (1) Order:Ack, (2) Ack:Tick, (3) Ref Prices. – How Fast Is Fast Enough? Millennium Average OK – Focus on Consistency. – Who really needs it? Predominantly Market Makers, HFTs, Co-Lo Algos. – What’s the next best thing? Consistency and Resilience. – The Technology • How did we get here? C++/Linux/Infiniband/Native/ITCH • What next? Balancing/threading/FPGA/Tickerplant, low level API/Generation 8 … • Performance & Capacity Management – The LSE/TQ KPIs Trans,Trades,latencies,TPS (1/10/60), Availability. – System behaviour under 1. Always respond, 2. Fail gracefully. extreme conditions 13