Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Updating frozen challenges #118

Draft
wants to merge 8 commits into
base: master
Choose a base branch
from
Draft

Updating frozen challenges #118

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
8 commits
Select commit Hold shift + click to select a range
ff9d238
Working on adding additional operations to the `frozen-data-generatio…
a03nikki Jun 11, 2024
6358261
Fixed typo in parameter definition for the challenge.
a03nikki Jun 11, 2024
22c517e
Fixed the delete and create ILM policy operations.
a03nikki Jun 11, 2024
3e08d1b
Added a note to the readme about license requirements for the `frozen…
a03nikki Jun 11, 2024
a4646f7
Create ILM policy does not take a file so removing the extra file.
a03nikki Jun 15, 2024
9a37392
Added missing challenge parameters for Frozen to the README.md file.
a03nikki Jul 3, 2024
329500d
Update the `frozen-querying` challenge to parameterize the number of …
a03nikki Jul 3, 2024
8f40463
Increase client count on the fieldstats call to match the search clie…
a03nikki Jul 5, 2024
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
34 changes: 30 additions & 4 deletions README.md
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -80,7 +80,7 @@ The table below shows the track parameters that can be adjusted along with defau

This challenge assumes that the *elasticlogs-1bn-load* track has been executed as it simulates querying against these indices. It shows how indexing and querying through simulated Kibana dashboards can be combined to provide a more realistic benchmark.

In this challenge rate-limited indexing at varying levels is combined with a fixed level of querying. If metrics from the run are stored in Elasticsearch, it is possible to analyse these in Kibana in order to identify how indexing rate affects query latency and vice versa.
In this challenge rate-limited indexing at varying levels is combined with a fixed level of querying. If metrics from the run are stored in Elasticsearch, it is possible to analyze these in Kibana in order to identify how indexing rate affects query latency and vice versa.

The table below shows the track parameters that can be adjusted along with default values:

Expand Down Expand Up @@ -173,7 +173,7 @@ Indexes several days of logs with a fixed (raw) logging volume per day and runni
| Parameter | Explanation | Type | Default Value |
| ----------------------- | -------------------------------------------------------------------------------------------------------------------------------------- | ----- | --------------------- |
| `bulk_indexing_clients` | Number of bulk indexing clients/connections | `int` | `8` |
| `search_clients` | Number of search clients/connections used by *each** query | `int` | `1` |
| `search_clients` | Number of search clients/connections used by **each** query | `int` | `1` |
| `bulk_size` | Number of documents to send per bulk | `int` | `1000` |
| `daily_logging_volume` | The raw logging volume. Supported units are bytes (without any unit), `KB`, `MB` and `GB`). For the value, only integers are allowed. | `str` | `100GB` |
| `starting_point` | The first timestamp for which logs should be generated. | `str` | `2018-05-25 00:00:00` |
Expand All @@ -188,8 +188,8 @@ Indexes (several days of) logs at a fixed target throughput using a fixed (raw)
| `bulk_indexing_reqs_per_sec` | Number of bulk indexing requests/sec. Multiply this by bulk_size to understand indexing throughput in docs/s. | `int` | `20` |
| `bulk_size` | Number of documents to send per bulk | `int` | `1000` |
| `bulk_indexing_clients` | Number of bulk indexing clients/connections | `int` | `8` |
| `search_clients` | Number of search clients/connections used by *each** query | `int` | `1` |
| `daily_logging_volume` | The raw logging volume. Supported units are bytes (without any unit), `KB`, `MB` and `GB`). For the value, only integers are allowed. | `str` | `100GB` |
| `search_clients` | Number of search clients/connections used by **each** query | `int` | `1` |
| `daily_logging_volume` | The raw logging volume. Supported units are bytes (without any unit), `KB`, `MB` and `GB`. For the value, only integers are allowed. | `str` | `100GB` |
| `starting_point` | The first timestamp for which logs should be generated. | `str` | `2018-05-25 00:00:00` |
| `number_of_days` | The number of simulated days for which data should be generated. | `int` | `6` |

Expand Down Expand Up @@ -237,6 +237,32 @@ This challenge has been used for the evaluation of query performance across diff
| `indices_recovery_max_bytes_per_sec` | If set, overrides Elasticsearch's default for [indices.recovery.max_bytes_per_sec](https://www.elastic.co/guide/en/elasticsearch/reference/current/recovery.html#recovery-settings) | `dict` | - |
| `query_max_concurrent_shard_requests` | If set, overrides Elasticsearch's default for [max_concurrent_shard_requests](https://www.elastic.co/guide/en/elasticsearch/reference/7.12/search-multi-search.html#search-multi-search-api-query-params) for the Kibana queries | `int` | - |

### frozen-data-generation

This challenge indexes data into an index that uses an ILM policy. After indexing is complete the index is manually rolled over to force it to be a frozen index.

This challenge requires a license to run as the default ILM policy uses the `searchable_snapshot` action. It also requires the snapshot repository to be setup ahead of time in the cluster.

| Parameter | Explanation | Type | Default Value |
|----------------------------|------------------------------------------------------------------------------------------------|--------|---------------|
| `bulk_indexing_iterations` | How many requests to send in total | `int` | `1000000` |
| `ilm_policy` | The name of the ILM policy to use. | `str` | `elasticlogs-hot-frozen-ilm-policy` |
| `es_snapshot_repo_name` | Used if `ilm_policy` is is not set. The name of the snapshot repository from which the snapshot should be stored | `str` | - |
| `frozen_min_age` | Used if `ilm_policy` is is not set. The minimum age after rollover an index is moved to frozen | `str` | `5s` |
| `rollover_max_shard_size` | Used if `ilm_policy` is is not set. Max primary shard size condition for [rollover API](https://www.elastic.co/guide/en/elasticsearch/reference/master/indices-rollover-index.html#indices-rollover-index) | `str` | `30gb` |
| `rollover_max_age` | Used if `ilm_policy` is is not set. Max age condition for [rollover API](https://www.elastic.co/guide/en/elasticsearch/reference/master/indices-rollover-index.html#indices-rollover-index) | `str` | `1d` |

### frozen-querying

This challenge queries frozen indices. It depends on data loaded by the `frozen-data-generation` challenge.

| Parameter | Explanation | Type | Default Value |
|---------------------------------|----------------------------------------------------------|--------|---------------|
| `max_concurrent_shard_requests` | Overrides Elasticsearch's default for [max_concurrent_shard_requests](https://www.elastic.co/guide/en/elasticsearch/reference/current/search-search.html#search-search-api-query-params) | `int` | `10` |
| `pre_filter_shard_size` | Overrides Elasticsearch's default for [`pre_filter_shard_size`](https://www.elastic.co/guide/en/elasticsearch/reference/current/search-search.html#search-search-api-query-params), used with throttled (frozen) indices. | `int` | `1` |
| `query_iterations` | Number of query requests per client | `int` | `3` |
| `search_clients` | Number of search clients/connections used for each query | `int` | `1` |

## Custom parameter sources

### elasticlogs\_bulk\_source
Expand Down
72 changes: 66 additions & 6 deletions eventdata/challenges/frozen.json
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -2,7 +2,8 @@
{% set p_pre_filter_shard_size = (pre_filter_shard_size | default(1)) %}
{% set p_bulk_idx_iterations = (bulk_indexing_iterations | default(30000)) %}
{% set p_query_iterations = (query_iterations | default(3)) %}

{% set p_ilm_policy = ( ilm_policy | default("elasticlogs-hot-frozen-ilm-policy")) %}
{% set p_search_clients = (search_clients | default(1)) %}
{
"name": "frozen-data-generation",
"description": "Index approximately 50GB of data into an index with with 2 primary shards. IDs are autogenerated by Elasticsearch, meaning there are no conflicts.",
Expand All @@ -13,11 +14,69 @@
{
"operation": "delete-index-template"
},
{% if p_ilm_policy == "elasticlogs-hot-frozen-ilm-policy" %}
{
"operation": {
"operation-type": "delete-ilm-policy",
"policy-name": "{{p_ilm_policy}}"
}
},
{
"operation": {
"operation-type": "create-ilm-policy",
"policy-name": "{{p_ilm_policy}}",
"body": {
"policy": {
"phases": {
"hot": {
"min_age": "0ms",
"actions": {
"rollover": {
"max_primary_shard_size": {{ rollover_max_shard_size | default("30gb") | tojson }},
"max_age": {{ rollover_max_age | default("1d") | tojson }}
},
"set_priority": {
"priority": 100
},
"forcemerge": {
"max_num_segments": 1
}
}
},
"frozen": {
"min_age": {{ frozen_min_age | default("5s") | tojson }},
"actions": {
"searchable_snapshot": {
"snapshot_repository": {{ es_snapshot_repo_name | default("-") | tojson }},
"force_merge_index": true
}
}
}
}
}
}
}
},
{%endif%}
{
"operation": {
"operation-type": "create-index-template",
"settings": {
"index.store.type": "hybridfs"
"index.store.type": "hybridfs",
"index.lifecycle.name": "{{p_ilm_policy}}",
"index.lifecycle.rollover_alias": "elasticlogs"
}
}
},
{
"operation": {
"name": "create-elasticlogs-write-alias",
"operation-type": "create-index",
"index": "{{p_index_prefix}}-000001",
"body": {
"aliases" : {
"elasticlogs" : { "is_write_index": true }
}
}
}
},
Expand Down Expand Up @@ -47,7 +106,7 @@
},
{
"name": "frozen-querying",
"description": "Frozen node querying.",
"description": "Frozen node querying. Depends on frozen-data-generation.",
"schedule": [
{
"operation": {
Expand All @@ -68,14 +127,15 @@
"iterations": 1,
"clients": 1
},

{
"operation": {
"operation-type": "fieldstats",
"index_pattern": "elasticlogs*",
"ignore_throttled": false
},
"iterations": 1,
"clients": 1
"clients": {{ p_search_clients }}
},
{
"name": "kibana-discover-ip-25%",
Expand All @@ -93,7 +153,7 @@
"pre_filter_shard_size": {{ p_pre_filter_shard_size }}
},
"iterations": {{ p_query_iterations }},
"clients": 1
"clients": {{ p_search_clients }}
},
{
"name": "kibana-content_issues-ip-25%",
Expand All @@ -111,7 +171,7 @@
"pre_filter_shard_size": {{ p_pre_filter_shard_size }}
},
"iterations": {{ p_query_iterations }},
"clients": 1
"clients": {{ p_search_clients }}
}
]
}