Performance and Exception Monitoring workshop HEPi X Braunschweig

Performance and Exception Monitoring workshop HEPi. X Braunschweig - 04/06/2000

"PEM" projects • • • Focus on • NGOP (Fermilab) • PEM (CERN) http: //proj-pem. cern. ch/proj-pem/ History and timescales Differences and similarities • • • http: //www-isd. fnal. gov/ngop/ Requirements Architecture Understanding the system Areas of common interest No implementation details yet. . . HEPi. X Braunschweig - 04/06/2000

Common history • Autumn 1999/Winter 2000 • February • March • Spring • User requirements, tools survey • Meeting at CHEP • Workshop and meeting at Fermilab • HEPi. X Braunschweig • NGOP/PEM meeting at CERN HEPi. X Braunschweig - 04/06/2000

Project timescales • PEM • • • May • End of analysis phase June • Design/protyping Summer • Implementation and deployement on selected clusters NGOP • May-August • Design • September-October • Testing • Winter 2001 • Integration and release • Time constraint • increasing number of PCs HEPi. X Braunschweig - 04/06/2000 • Run II

User requirements • • Mostly the same Different focus on operators • • • NGOP - and OPEP 2 - include host management tools PEM allows only automatic recovery actions, alarm severities not built-in Scope • • Both include only partially the implementation of monitoring agents PEM does not include user interfaces, although it will provide some HEPi. X Braunschweig - 04/06/2000

Architecture • NGOP • • • PEM • • • based on exception monitoring performance monitoring added separately based on performance monitoring exception monitoring added on top Both foresee • • events/measurements that cannot be generated locally metric hierarchies HEPi. X Braunschweig - 04/06/2000

Metric hierarchy 1. PEM glossary 1. Service 1. A predefined set of functionalities provided to users on a set of hosts 2. NGOP equivalent: cluster 2. Host 3. Host type 4. Metric (simple or composite) 1. A computing equipment with a network interface 1. The property of a set of hosts of having a common goal 1. From GQM 2. NGOP equivalent: system/subsystem/compenent HEPi. X Braunschweig - 04/06/2000

PEM Framework Diagram HEPi. X Braunschweig - 04/06/2000

PEM sub-systems Monitoring Client Monitoring Server Configuration DB Alarm GUI Notifier History Display Report Generator Access Server HEPi. X Braunschweig - 04/06/2000 Measurement DB Correlation Server

NGOP sub-systems Sensor Component DBServer Monitoring Agent Name Server Persis. Config Data Alarm GUI Notifier History Display Report Generator HEPi. X Braunschweig - 04/06/2000 Central Server Looping Mon Agent

What to measure • What quantities have to be monitored? • Use GQM • Goal to be achieved • How to achieve it • What to measure in order to verify • Example Goal Question Metric Provide information about machine configuration What is the hardware configuration Number, type and clock of CPU(s) HEPi. X Braunschweig - 04/06/2000

Discussion (1) • Areas of collaboration • • • GQM Configuration Correlation language and engine Interface to repositories Agents to collect information HEPi. X Braunschweig - 04/06/2000

Discussion (2) • Open points • • • Do we need different glossaries? Do we need multiple architectures? Role of operators Role of system openness Contacts tim. smith@cern. ch alessandro. miotto@cern. ch HEPi. X Braunschweig - 04/06/2000