Migration Guide

Arc Cloud supports bulk import of CSV, Parquet, and line protocol files, making it straightforward to migrate data from InfluxDB or any other source. The import endpoints accept files up to 500 MB per request and support gzip compression.

Throughout this guide, replace the base URL with your instance's URL:

https://<instance-id>.arc.<region>.basekick.net

From InfluxDB

1. Export Data as Line Protocol

Use the influx CLI to export your data in line protocol format:

influx -database mydb -execute "SELECT * FROM cpu" -format lp > cpu_export.lp

For large datasets, export one measurement at a time or add a WHERE time > ... clause to split by time range.

2. Import into Arc Cloud

curl -X POST "https://<instance-id>.arc.<region>.basekick.net/api/v1/import/lp?db=mydb" \
  -H "Authorization: Bearer <token>" \
  -H "Content-Type: text/plain" \
  --data-binary @cpu_export.lp

With gzip compression (recommended for large files):

gzip cpu_export.lp

curl -X POST "https://<instance-id>.arc.<region>.basekick.net/api/v1/import/lp?db=mydb" \
  -H "Authorization: Bearer <token>" \
  -H "Content-Type: text/plain" \
  -H "Content-Encoding: gzip" \
  --data-binary @cpu_export.lp.gz

3. Verify

curl -X POST https://<instance-id>.arc.<region>.basekick.net/api/v1/query \
  -H "Authorization: Bearer <token>" \
  -H "Content-Type: application/json" \
  -d '{"q": "SELECT count(*) FROM cpu"}'

From CSV

Arc Cloud can import CSV files directly. You need to specify the measurement name and which column contains the timestamp.

Prepare Your CSV

Ensure your CSV has a header row and a parseable timestamp column:

timestamp,host,region,usage,temperature
2026-03-23T12:00:00Z,server01,us-east,0.64,72.1
2026-03-23T12:01:00Z,server02,eu-west,0.38,68.4

Import

curl -X POST "https://<instance-id>.arc.<region>.basekick.net/api/v1/import/csv?db=mydb&measurement=sensors&time_column=timestamp" \
  -H "Authorization: Bearer <token>" \
  -H "Content-Type: text/csv" \
  --data-binary @data.csv

Parameters

Parameter	Description	Default
db	Target database	required
measurement	Target measurement name	required
time_column	Name of the timestamp column	timestamp
delimiter	Column delimiter	,
tag_columns	Comma-separated list of columns to treat as tags	none (all columns are fields)

Example with Custom Delimiter and Tags

curl -X POST "https://<instance-id>.arc.<region>.basekick.net/api/v1/import/csv?db=mydb&measurement=sensors&time_column=ts&delimiter=;&tag_columns=host,region" \
  -H "Authorization: Bearer <token>" \
  -H "Content-Type: text/csv" \
  --data-binary @data.csv

From Parquet

Parquet is ideal for migrating large datasets. The columnar format is compact and fast to import.

curl -X POST "https://<instance-id>.arc.<region>.basekick.net/api/v1/import/parquet?db=mydb&measurement=sensors" \
  -H "Authorization: Bearer <token>" \
  -H "Content-Type: application/octet-stream" \
  --data-binary @data.parquet

The Parquet file should include a timestamp column. Arc will auto-detect common timestamp column names (time, timestamp, ts, datetime). To specify explicitly, add &time_column=your_column to the query string.

Tips for Large Migrations

• Split large exports into files under 500 MB each. For line protocol exports, split by measurement or time range.
• Use gzip compression to reduce transfer time. All import endpoints accept Content-Encoding: gzip.
• Import in parallel -- Arc Cloud handles concurrent imports safely. You can run multiple curl commands simultaneously for different measurements.
• Verify row counts after import by querying SELECT count(*) FROM <measurement> and comparing against the source.