EPICS Experimental Physics and Industrial Control System Control

  • Slides: 19
Download presentation
EPICS: Experimental Physics and Industrial Control System Control Architecture Reading Group

EPICS: Experimental Physics and Industrial Control System Control Architecture Reading Group

Overview w What, Why and Who? w The Subsystems w Performance w Conclusions June

Overview w What, Why and Who? w The Subsystems w Performance w Conclusions June 23, 2004 Control Architecture Reading Group

What is EPICS and Why? w Scaleable “real-time” remote control – distributed systems –

What is EPICS and Why? w Scaleable “real-time” remote control – distributed systems – small test stands w Client / Server Model – Server: low-level hardware – Client: user interface w w w Control: supervisory, closed-loop and sequential “Configuration tools in place of programming” Large installed base of tested software Modular design that supports incremental upgrades Well defined interfaces for extensions at every level June 23, 2004 Control Architecture Reading Group

Who is Using EPICS? w Over 90 independent projects in North America, Europe and

Who is Using EPICS? w Over 90 independent projects in North America, Europe and Asia – – – – Los Alamos National Laboratory Argonne National Laboratory Lawrence Berkeley Laboratory Superconducting Super Collider Laboratory Continuous Electron Beam Accelerator Facility University of Saskatchewan, UBC Duke University, Stanford Scientific Instrument Limited June 23, 2004 Control Architecture Reading Group

Accelerators: Think BIG! June 23, 2004 Control Architecture Reading Group

Accelerators: Think BIG! June 23, 2004 Control Architecture Reading Group

EPICS Subsystems (1) June 23, 2004 Control Architecture Reading Group

EPICS Subsystems (1) June 23, 2004 Control Architecture Reading Group

EPICS Subsystems (2) June 23, 2004 Control Architecture Reading Group

EPICS Subsystems (2) June 23, 2004 Control Architecture Reading Group

Distributed Database (Servers) w Database: local control – Highest level on each IOC –

Distributed Database (Servers) w Database: local control – Highest level on each IOC – Above hardware drivers – Simple config file w w w Data Acquisition Data Conversion Alarm Detection Closed Loop Control 4 -100 k. Hz PID loops June 23, 2004 Control Architecture Reading Group

Display Manager (Client) w Interface to Operator w X-Windows w Strip Charts, etc. June

Display Manager (Client) w Interface to Operator w X-Windows w Strip Charts, etc. June 23, 2004 Control Architecture Reading Group

Alarm Manager (Client) w “Fault Trees” w Steady State Operation w Give guidance to

Alarm Manager (Client) w “Fault Trees” w Steady State Operation w Give guidance to operator June 23, 2004 Control Architecture Reading Group

Archiver (Client) w Data to Disk w Select Channels to Retrieve w 5000 Channels

Archiver (Client) w Data to Disk w Select Channels to Retrieve w 5000 Channels / sec w Multiple Archivers at once on network June 23, 2004 Control Architecture Reading Group

Sequencer (Client) w Execute State Machines w Runs on each IOC w “State Notation

Sequencer (Client) w Execute State Machines w Runs on each IOC w “State Notation Language” w Switches Op. Modes w Handles Exceptions w C code can be added June 23, 2004 Control Architecture Reading Group

Channel Access w Controls how clients and servers talk to each other w “Software

Channel Access w Controls how clients and servers talk to each other w “Software Bus” w Over TCP or UDP w Establish connections w Get, Put, Monitor Info June 23, 2004 Control Architecture Reading Group

Event Synchronization w “Real Time” across network – millisecond time-stamps w Measure same event

Event Synchronization w “Real Time” across network – millisecond time-stamps w Measure same event across network w Based on individual local machine clock w Avoid Ethernet “collisions” June 23, 2004 Control Architecture Reading Group

I/O & Network Performance w 4 -100 k. Hz IOC low-level loops w <

I/O & Network Performance w 4 -100 k. Hz IOC low-level loops w < 60 Hz Channel Access Loops w 10, 000 Channel Access monitors per second on 10 MBit Ethernet w Ethernet load < 30% (for determinism) w Signal latency on network: 2 ms+ – 68040 on 10 Mbit Ethernet w “Network bandwidth is the primary limiting factor” w The Ground Test Accelerator (old stats) – – 2, 500 physical connections 10, 000 database records in 14 IOCs 8 workstations 5 -7% of 10 Mbit Ethernet. June 23, 2004 Control Architecture Reading Group

Reliability w Accidents are Expensive! w 95% uptime w Lots of testing, been around

Reliability w Accidents are Expensive! w 95% uptime w Lots of testing, been around for years w Not a toy or pet project June 23, 2004 Control Architecture Reading Group

Portability w Vx. Works, Linux, Windows, RTEMS, Darwin, Solaris w Control Net, PCI, CAN-Bus,

Portability w Vx. Works, Linux, Windows, RTEMS, Darwin, Solaris w Control Net, PCI, CAN-Bus, Industry Pack, VME, VXI, PCI, ISA, CAMAC, GPIB, Profibus, Bitbus, Serial, Allen. Bradley, Modbus, Yokogawa, G-3, Ethernet/IP w 500 k. B+ Server Executable w RTEMS vs. Vx. Works – IOC – Critical: Hard real-time – RTEMS as fast as Vx. Works w Linux (“all-in-one”) – non-critical systems June 23, 2004 Control Architecture Reading Group

Conclusion w Scaleable w Distributed w Deterministic & relatively fast – (1 ms time-stamps)

Conclusion w Scaleable w Distributed w Deterministic & relatively fast – (1 ms time-stamps) w Reliable w Ethernet-based control architecture w Standard open-source Unix tools June 23, 2004 Control Architecture Reading Group

Further Reading w “Recommended” Documents w http: //lansce. lanl. gov/lansce 8/Epics/epi cs. X 5

Further Reading w “Recommended” Documents w http: //lansce. lanl. gov/lansce 8/Epics/epi cs. X 5 Farch-1. html w EPICS Architecture @ ANL w EPICS: Recent Developments and Future Perspectives w EPICS on the RTEMS real-time executive for multiprocessor systems June 23, 2004 Control Architecture Reading Group