Partitioning

Whenever possible, old temporary entries are deleted through a sliding window procedure. However for many tables the information is preferred to be kept over extended periods of time (years) in order to be able to browse the data in the monitoring. This leads to very large databases with many million rows. In addition to the usage of several indexes, Oracle partitioning is also used to optimize data access. Below are the current partitioning strategies and how new partitions are added (automatically, in a procedure or manually).

  • ATLAS_PANDA

Table

Partitioned by

Range

Handled

Datasets

modificationtime

1 month

AUTO

Filestable4

modificationtime

1 day

ADD_DAILYPART proc

Harvester_dialogs

creationtime

1 day

AUTO

Harvester_metrics

creation_time

1 day

AUTO

Harvester_rel_jobs_workers

pandaid

1E8

AUTO

Harvester_workers

lastupdate

1 month

AUTO

Jedi_datasets

Jeditaskid

500k

AUTO

Jedi_dataset_contents

Jeditaskid

500k

AUTO

Jedi_events

Jeditaskid

500k

MANUAL

Jedi_jobparams_template

Ins_UTC_TSTAMP

1 day

AUTO

Jedi_output_template

Jeditaskid

500k

AUTO

Jedi_taskparams

Jeditaskid

500k

AUTO

Jedi_tasks

Jeditaskid

500k

AUTO

JOBPARAMSTABLE

MODIFICATIONTIME

1 day

ADD_DAILYPART proc

JOBSARCHIVED4

MODIFICATIONTIME

1 day

ADD_DAILYPART proc

JOBS_STATUSLOG

MODIFICATIONTIME

1 day

AUTO

METATABLE

MODIFICATIONTIME

1 day

ADD_DAILYPART proc

PANDALOG

BINTIME

1 day

AUTO

PANDALOG_FAX

BINTIME

1 month

AUTO

TABLEPART4COPYING

COPIED_TO_ARCH

values

STATIC

TASKS_STATUSLOG

MODIFICATIONTIME

1 day

AUTO

TASKS_STATUSLOG

MODIFICATIONTIME

1 day

AUTO

  • ATLAS_PANDAARCH

Table

Partitioned by

Range

Handled

FILESTABLE_ARCH

MODIFICATIONTIME

1 month

MANUAL

JOBPARAMSTABLE_ARCH

MODIFICATIONTIME

1 month

MANUAL

JOBSARCHIVED

MODIFICATIONTIME

3 days

MANUAL

METATABLE_ARCH

MODIFICATIONTIME

1 month

MANUAL