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

Jump to bottom

Add metrics for CQL request and response sizes #13061

Closed
vladzcloudius opened this issue Mar 2, 2023 · 6 comments
Closed

Add metrics for CQL request and response sizes #13061

vladzcloudius opened this issue Mar 2, 2023 · 6 comments
Labels
area/cql area/monitoring tracing feature/enhancement
Milestone
5.3

Comments

@vladzcloudius
Copy link
Contributor

vladzcloudius commented Mar 2, 2023 •
edited

Installation details
HEAD: b6e4275

Description
We need metrics that would allow us reasoning about sizes of requests and the amount of data that is returned.

We can start with averages but histograms would definitely make sense eventually.
vladzcloudius added a commit to vladzcloudius/scylla that referenced this issue Mar 2, 2023
@vladzcloudius
transport: refactor CQL metrics
b7c4147 
This patch reorganizes and extends CQL related metrics.
Before this patch we only had counters for specific CQL requests.

However, many times we need to reason about the size of CQL queries: corresponding
requests and response sizes.

This patch adds corresponding metrics:
  - Arranges all 3 per-opcode statistics counters in a single struct.
  - Defines a vector of such structs for each CQL opcode.
  - Adjusts statistics updates accordingly - the code is much simpler
    now.
  - Removes old metrics that were accounting some CQL opcodes.
  - Adds new per-opcode metrics for requests number, request and response sizes:
     - New metrics are of a derived kind - rate() should be applied to them.
     - There are 3 new metrics names:
       - 'cql_requests_count'
       - 'cql_request_bytes'
       - 'cql_response_bytes'
     - New metrics have a per-opcode label - 'kind'.

 For example:

 A number of response bytes for an EXECUTE opcode on shard 0 looks as follows:

 scylla_transport_cql_response_bytes{kind="EXECUTE",shard="0"}

Ref scylladb#13061

Signed-off-by: Vlad Zolotarov <vladz@scylladb.com>
vladzcloudius added a commit to vladzcloudius/scylla that referenced this issue Mar 2, 2023
@vladzcloudius
transport: refactor CQL metrics
225611a 
This patch reorganizes and extends CQL related metrics.
Before this patch we only had counters for specific CQL requests.

However, many times we need to reason about the size of CQL queries: corresponding
requests and response sizes.

This patch adds corresponding metrics:
  - Arranges all 3 per-opcode statistics counters in a single struct.
  - Defines a vector of such structs for each CQL opcode.
  - Adjusts statistics updates accordingly - the code is much simpler
    now.
  - Removes old metrics that were accounting some CQL opcodes.
  - Adds new per-opcode metrics for requests number, request and response sizes:
     - New metrics are of a derived kind - rate() should be applied to them.
     - There are 3 new metrics names:
       - 'cql_requests_count'
       - 'cql_request_bytes'
       - 'cql_response_bytes'
     - New metrics have a per-opcode label - 'kind'.

 For example:

 A number of response bytes for an EXECUTE opcode on shard 0 looks as follows:

 scylla_transport_cql_response_bytes{kind="EXECUTE",shard="0"}

Ref scylladb#13061

Signed-off-by: Vlad Zolotarov <vladz@scylladb.com>
@mykaul mykaul added this to the 5.3 milestone Mar 5, 2023
avikivity pushed a commit that referenced this issue Mar 7, 2023
@vladzcloudius @avikivity
transport: refactor CQL metrics
ae6724f 
This patch reorganizes and extends CQL related metrics.
Before this patch we only had counters for specific CQL requests.

However, many times we need to reason about the size of CQL queries: corresponding
requests and response sizes.

This patch adds corresponding metrics:
  - Arranges all 3 per-opcode statistics counters in a single struct.
  - Defines a vector of such structs for each CQL opcode.
  - Adjusts statistics updates accordingly - the code is much simpler
    now.
  - Removes old metrics that were accounting some CQL opcodes.
  - Adds new per-opcode metrics for requests number, request and response sizes:
     - New metrics are of a derived kind - rate() should be applied to them.
     - There are 3 new metrics names:
       - 'cql_requests_count'
       - 'cql_request_bytes'
       - 'cql_response_bytes'
     - New metrics have a per-opcode label - 'kind'.

 For example:

 A number of response bytes for an EXECUTE opcode on shard 0 looks as follows:

 scylla_transport_cql_response_bytes{kind="EXECUTE",shard="0"}

Ref #13061

Signed-off-by: Vlad Zolotarov <vladz@scylladb.com>
Message-Id: <20230302154816.299721-1-vladz@scylladb.com>
avikivity pushed a commit that referenced this issue Mar 22, 2023
@vladzcloudius @avikivity
transport: add per-scheduling-group CQL opcode-specific metrics
f94bbc5 
This patch extends a previous patch that added these metrics globally:
 - cql_requests_count
 - cql_request_bytes
 - cql_response_bytes

This patch adds a "scheduling_group_name" label to these metrics and changes corresponding
counters to be accounted on a per-scheduling-group level.

As a bonus this patch also marks all 3 metrics as 'skip_when_empty'.

Ref #13061

Signed-off-by: Vlad Zolotarov <vladz@scylladb.com>
Message-Id: <20230321201412.3004845-1-vladz@scylladb.com>
vladzcloudius added a commit to vladzcloudius/scylla that referenced this issue Mar 23, 2023
@vladzcloudius
transport: refactor CQL metrics
2b6f45d 
This patch reorganizes and extends CQL related metrics.
Before this patch we only had counters for specific CQL requests.

However, many times we need to reason about the size of CQL queries: corresponding
requests and response sizes.

This patch adds corresponding metrics:
  - Arranges all 3 per-opcode statistics counters in a single struct.
  - Defines a vector of such structs for each CQL opcode.
  - Adjusts statistics updates accordingly - the code is much simpler
    now.
  - Removes old metrics that were accounting some CQL opcodes.
  - Adds new per-opcode metrics for requests number, request and response sizes:
     - New metrics are of a derived kind - rate() should be applied to them.
     - There are 3 new metrics names:
       - 'cql_requests_count'
       - 'cql_request_bytes'
       - 'cql_response_bytes'
     - New metrics have a per-opcode label - 'kind'.

 For example:

 A number of response bytes for an EXECUTE opcode on shard 0 looks as follows:

 scylla_transport_cql_response_bytes{kind="EXECUTE",shard="0"}

Ref scylladb#13061

Signed-off-by: Vlad Zolotarov <vladz@scylladb.com>
Message-Id: <20230302154816.299721-1-vladz@scylladb.com>
(cherry picked from commit ae6724f)
vladzcloudius added a commit to vladzcloudius/scylla that referenced this issue Mar 23, 2023
@vladzcloudius
transport: add per-scheduling-group CQL opcode-specific metrics
6537306 
This patch extends a previous patch that added these metrics globally:
 - cql_requests_count
 - cql_request_bytes
 - cql_response_bytes

This patch adds a "scheduling_group_name" label to these metrics and changes corresponding
counters to be accounted on a per-scheduling-group level.

As a bonus this patch also marks all 3 metrics as 'skip_when_empty'.

Ref scylladb#13061

Signed-off-by: Vlad Zolotarov <vladz@scylladb.com>
Message-Id: <20230321201412.3004845-1-vladz@scylladb.com>
(cherry picked from commit f94bbc5)
@vladzcloudius vladzcloudius mentioned this issue Mar 23, 2023
Closed
@DoronArazii
Copy link

@mykaul / @vladzcloudius can we close it? or it requires more work?

@vladzcloudius
Copy link
Contributor Author

@mykaul / @vladzcloudius can we close it? or it requires more work?

We should close it only after histograms are added (mentioned in the opening message).

@mykaul
Copy link
Contributor

mykaul commented Apr 20, 2023

@mykaul / @vladzcloudius can we close it? or it requires more work?

We should close it only after histograms are added (mentioned in the opening message).

Thanks @vladzcloudius - will you be able to complete this work (we hope to branch out 5.3 very soon). Alternatively, we may prefer to open a separate GH issue to track that activity?

@vladzcloudius
Copy link
Contributor Author

@mykaul / @vladzcloudius can we close it? or it requires more work?

We should close it only after histograms are added (mentioned in the opening message).

Thanks @vladzcloudius - will you be able to complete this work (we hope to branch out 5.3 very soon). Alternatively, we may prefer to open a separate GH issue to track that activity?

No. I'm not planning to work on that any time soon.

@mykaul
Copy link
Contributor

mykaul commented Apr 20, 2023

@mykaul / @vladzcloudius can we close it? or it requires more work?

We should close it only after histograms are added (mentioned in the opening message).

Thanks @vladzcloudius - will you be able to complete this work (we hope to branch out 5.3 very soon). Alternatively, we may prefer to open a separate GH issue to track that activity?

No. I'm not planning to work on that any time soon.

Thanks - then I'm in favor of opening a separate item to track the histogram request.

@DoronArazii
Copy link

@mykaul / @vladzcloudius can we close it? or it requires more work?

We should close it only after histograms are added (mentioned in the opening message).

Closing as completed.

@vladzcloudius vladzcloudius mentioned this issue Jul 26, 2023
Open
kbr-scylla pushed a commit to kbr-scylla/scylladb that referenced this issue Aug 2, 2023
@vladzcloudius
transport: refactor CQL metrics
cd1a1b9 
This patch reorganizes and extends CQL related metrics.
Before this patch we only had counters for specific CQL requests.

However, many times we need to reason about the size of CQL queries: corresponding
requests and response sizes.

This patch adds corresponding metrics:
  - Arranges all 3 per-opcode statistics counters in a single struct.
  - Defines a vector of such structs for each CQL opcode.
  - Adjusts statistics updates accordingly - the code is much simpler
    now.
  - Removes old metrics that were accounting some CQL opcodes.
  - Adds new per-opcode metrics for requests number, request and response sizes:
     - New metrics are of a derived kind - rate() should be applied to them.
     - There are 3 new metrics names:
       - 'cql_requests_count'
       - 'cql_request_bytes'
       - 'cql_response_bytes'
     - New metrics have a per-opcode label - 'kind'.

 For example:

 A number of response bytes for an EXECUTE opcode on shard 0 looks as follows:

 scylla_transport_cql_response_bytes{kind="EXECUTE",shard="0"}

Ref scylladb#13061

Signed-off-by: Vlad Zolotarov <vladz@scylladb.com>
Message-Id: <20230302154816.299721-1-vladz@scylladb.com>
(cherry picked from commit ae6724f)
kbr-scylla pushed a commit to kbr-scylla/scylladb that referenced this issue Aug 2, 2023
@vladzcloudius
transport: add per-scheduling-group CQL opcode-specific metrics
8f1cf11 
This patch extends a previous patch that added these metrics globally:
 - cql_requests_count
 - cql_request_bytes
 - cql_response_bytes

This patch adds a "scheduling_group_name" label to these metrics and changes corresponding
counters to be accounted on a per-scheduling-group level.

As a bonus this patch also marks all 3 metrics as 'skip_when_empty'.

Ref scylladb#13061

Signed-off-by: Vlad Zolotarov <vladz@scylladb.com>
Message-Id: <20230321201412.3004845-1-vladz@scylladb.com>
(cherry picked from commit f94bbc5)
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Assignees
No one assigned
Labels
area/cql area/monitoring tracing feature/enhancement
Projects
None yet
Milestone
5.3
Development

No branches or pull requests
3 participants
@mykaul @vladzcloudius @DoronArazii