Management of the LHCb DAQ Network Guoming Liu *† , Niko Neufeld * * CERN, Switzerland † University of Ferrara, Italy
Jan 04, 2016
Management of the LHCb DAQ Network
Guoming Liu*†, Niko Neufeld*
* CERN, Switzerland† University of Ferrara, Italy
2
Outline
Introduction to LHCb DAQ system
Network Monitoring based on SCADA system
Network Configuration
Network Debugging
Status of LHCb network installation and deployment
3
LHCb online system
LHCb Online system consists of three major components:
Data Acquisition (DAQ) transfers the event data from the detector front-end electronics
to the permanent storage
Timing and Fast Control (TFC) drives all stages of the data readout of the LHCb detector
between the front-end electronics and the online processing farm
Experiment Control System (ECS), controls and monitors all parts of the experiment: the DAQ
System, the TFC Systems, the High Level Trigger Farm, the Detector Control System, the Experiment's Infrastructure etc.
4
LHCb online system
Control and Monitoring data
CASTOR
SWITCH
HLT farm
Detector
TFC System
SWITCHSWITCH SWITCH SWITCH SWITCH SWITCH
READOUT NETWORK
L0 triggerLHC clock
MEP Request
Event building
Front-End
CPU
CPU
CPU
CPU
CPU
CPU
CPU
CPU
CPU
CPU
CPU
CPU
CPU
CPU
CPU
CPU
CPU
CPU
CPU
CPU
CPU
CPU
CPU
CPU
Readout Board
Expe
rimen
t Con
trol
Sys
tem
(EC
S)
VELO ST OT RICH ECal HCal Muon
L0 Trigger
Event dataTiming and Fast Control Signals
SWITCH
MON farm
CPU
CPU
CPU
CPU
Readout Board
Readout Board
Readout Board
Readout Board
Readout Board
Readout Board
FEE FEE FEE FEE FEE FEE FEE
5
LHCb online network
Two large scale Ethernet networks: DAQ network Dedicated to data
acquisition Control network For the instruments and
computers in LHCb experiment
In total: ~170 switches ~9000 ports
TELL1 ccpc switches
sw-d3b07-c1TFC
sw-d3b07-d1
Dummy det.sw-d2e01-d1
sw-daq-01
sw-agg-01
sw-agg-02
HLT FARMsw-d1exx-d1
HLT FARMsw-d1axx-d1
sw-d1exx-c1
sw-d1axx-c1
sw-d2e01-c1
sw-ux-01
L0Muon Trgsw-d3a01-01Sw-d3a03-01
sw-d2c08-01instruments
sw-d2c05-01
sw-d2c05-02
sw-d2d05-01
Storage Aggregationsw-d2b07-s1
sw-d2b05-s1
sw-d2b01-s1
sw-d2a07-d1
sw-d2a08-d1
DAQCONTROL
MONITORING
UKL1RICH1: sw-d3c01-01RICH2: sw-d3c04-01
sw-d2a07-c1
sw-d2a08-c1
sw-d2a08-01
sw-d2c05-m1IP: 10.132.10.21
10 G
DATA AGGREGATION
STORAGEsw-d2b04-s1sw-d2b03-s1sw-d2b02-s1
sw-d1dxx-c1sw-d2cxx-c1sw-d2bxx-c1
sw-d1dxx-c1sw-d2cxx-c1sw-d2bxx-c1
sw-d2d07-c1Calibration Farm
6
LHCb DAQ network
DAQ works in a push mode
Components: Readout board:
TELL1/UKL1 In total: ~330
Aggregation switches Core DAQ switch: Force10 E1200i
Supports up to 1260 GbE ports
Switch capacity: 3.5Tb/s Edge switches
Core Switch
HLT CPU
50 EdgeSwitches
~330 Readout Boards
HLT CPUHLT CPU
Storage Aggregation
CASTOR
AggregationSwitches
7
LHCb DAQ network
Core Switch
HLT CPU
50 EdgeSwitches
~330 Readout Boards
HLT CPUHLT CPU
Storage Aggregation
CASTOR
AggregationSwitches
Protocols Readout: MEP light-weight datagram
protocol over IP Storage: standard TCP/IP
Network throughputs Read out: ~35 GByte/s L0 trigger accept rate: 1
MHz Avg. event size: ~ 35 kByte
Storage: ~ 70 MByte/s HLT accept rate: ~ 2 kHz
~280 Gb/s
~560 Mb/s
8
Network Monitoring
Part of the LHCb ECS Uses the same tool and
framework Provides the same
operation interface
Implementation Monitoring and
integration: PVSS and JCOP Data collection: Varied front-end
processors Data exchange: Distributed Information
Management (DIM)
9
Architecture of the Network Monitoring
Network Monitoring
Monitoring the status of the LHCb DAQ network at different levels Topology IP routing Traffic Hardware/system
10
Network Monitoring
Monitoring the status of the LHCb DAQ network at different levels Topology IP routing Traffic Hardware/system
Structure of the Finite State Machinefor Network Monitoring
11
Network Monitoring: Topology
The topology is quite “static” NeDi: an open source tool to discover the network
Discovery of the network topology based on Link Layer Discovery Protocol (LLDP)
Queries the neighbors of the seed, and then the neighbors of those neighbors, and so on until all the devices have been discovered in the network.
Discovery of the network nodes
All information is stored in the database, and can be queried by PVSS
PVSS Monitors the topology only (the uplinks between the switches). The nodes are monitored by Nagios.
12
Network Monitoring: IP routing
Monitoring the status of the routing with Internet Control Message Protocol (ICMP), specifically “ping“
Three stages for the DAQ: Entire read-out event from the readout board to HLT farm ICMP not fully implemented in the readout board, a general
computer is inserted to simulate the the readout board: Test the status of the readout board using “arping” Test the availability of the HLT nodes using “ping”
Selected events from the HLT to the LHCb online storage From the online storage to CERN CASTOR
The front-end script gets the result and exchanges the message with PVSS using DIM
13
Network Monitoring: traffic
Front-end processors: Collect all the interface counters from the network devices
using SNMP Input and output traffic Input and output errors, discards
Exchange data as a DIM server
PVSS: Receives the data via PVSS-DIM bridge Analyzes the traffic and archives them Displays the current status and trending of the bandwidth
utilization Issues alarm in case of error
14
Network Monitoring: traffic
15
Network Monitoring: hardware/system
Syslog server is setup to receive the syslog messages from the network devices and parse the messages.
When the network devices run into problems, the error messages will be generated and sent to the syslog server as configured in the network device Hardware: temperature, fan status, power supply status System: CPU, memory, login authentication etc.
Syslog can collect some information not covered by SNMP
All the collected messages will be communicated to PVSS
16
Network Configuration
The LHCb online network system is quite large: Different devices with different OS and command sets But quite static luckily, only a few features are essential for
configuring the network devices.
Currently a set of Python scripts is used for configuring the network devices, using module pexpect for interactive CLI access. Initial setup for new installed switch Firmware upgrade Configuration file backup and restore
17
Network Configuration
NeDi CLI access
Web-based interface
Possible to select a set of switches by type, IP, or name etc.
Can apply a batch of commands on a set of switches
18
Network Diagnostics Tools
sFlow Sampler sFlow is a mechanism to capture packet headers, and collect
the statistics from the device, especially in high speed networks
samples the packet on the switch port and displays the header information
It is very useful to debug the packet loss problem, e.g. caused by wrong IP or MAC address
Relative high speed traffic monitoring Queries the counters for selected interfaces using SNMP or
CLI with a finer time resolution Shows the utilization for the selected interfaces
19
Status of Network Installation and Deployment
Current setup: With 2 aggregation switches Only 2 linecards inserted to the core DAQ switch For L0 trigger rate at ~200kHz
Upgrade for 1 MHz full speed readout. Core DAQ switch: Forec10 E1200i
14 linecards, 1260 GbE ports will be ready at the end of June Upgrade from Terascale to Exascale: double the switch capacity
and all ports run in line rate All readout boards will be connected to the core DAQ switch
directly
20