Skip to content

Run full planet #1323

@ivanPyrohivskyi

Description

@ivanPyrohivskyi

It is continue of issue #1322

My trying of start on the "weak" server.
Trying with use shared memory:
./osrm-datastore /mnt/data/planet_data/planet-latest.osrm
[info] load names from: "/mnt/data/planet_data/planet-latest.osrm.names"
[info] name offsets size: 313832
[info] allocating shared memory of 27864662738 bytes
Killed

Trying run osrm-routed:
./osrm-routed
[info] Reading options from: server.ini
[info] starting up engines, v0.4.3, compiled at Oct 16 2014, 16:59:33
[info] HSGR file: "/mnt/data/planet_data/planet-latest.osrm.hsgr"
[info] loading graph data
[info] loading graph from /mnt/data/planet_data/planet-latest.osrm.hsgr
[info] number_of_nodes: 216837848, number_of_edges: 1211389093
[info] loaded 216837848 nodes and 1211389093 edges
[info] Data checksum is 1576355079
[info] loading egde information
[info] loading geometries
Killed

Why process was killed?

Free RAM ~22Gb, swap file on SSD 90Gb.
On other (external) server full planet process got 28.4Gb of RAM.

If I increase RAM on the "weak" server to 32GB (now 24Gb, ~22Gb is free), will be it enough for run full planet routing?

Metadata

Metadata

Assignees

No one assigned

    Labels

    No labels
    No labels

    Type

    No type

    Projects

    No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions