Presto连接SQL Server查询响应缓慢问题求助

Hey everyone, I'm hitting a really confusing performance bottleneck with the Presto SQL Server connector. I'm running a basic SELECT query against a temp table that only has 2 rows, and the total time (planning + execution) in Presto is over 20 seconds. But when I run the exact same query directly in SQL Server, it completes in less than 50ms. I've got screenshots from the Presto dashboard that show this behavior, though I can't embed them here right now.

Here's my setup:

SQL Server Connector Config #

connector.name=sqlserver
connection-url=jdbc:sqlserver://[server address]
connection-user=user
connection-password=password

Presto Coordinator Settings #

coordinator=true
node-scheduler.include-coordinator=true
http-server.http.port=8080
query.max-memory=5GB
query.max-memory-per-node=1GB
discovery-server.enabled=true
discovery.uri=http://localhost:8080

And here's the output from EXPLAIN ANALYZE for the slow query:

Fragment 1 [SOURCE]
CPU: 10.87ms, Input: 2 rows (44B); per task: avg.: 2.00 std.dev.: 0.00, Output: 2 rows (44B)
Output layout: [id, col1]
Output partitioning: SINGLE []
Execution Flow: UNGROUPED_EXECUTION
- TableScan[sqlserver:sqlserver:dbo.temp:sns_viz:dbo:temp, originalConstraint = true] => [id:char(10), col1:varchar]
Cost: {rows: ? (?), cpu: ?, memory: 0.00, network: 0.00}
CPU fraction: 100.00%, Output: 2 rows (44B)
Input avg.: 2.00 rows, Input std.dev.: 0.00%
id := JdbcColumnHandle{connectorId=sqlserver, columnName=id, jdbcTypeHandle=JdbcTypeHandle{jdbcType=-15, columnSize=10, decimalDigits=0}, columnType=char(10)}
col1 := JdbcColumnHandle{connectorId=sqlserver, columnName=col1, jdbcTypeHandle=JdbcTypeHandle{jdbcType=-9, columnSize=2147483647, decimalDigits=0}, columnType=varchar}

The weird thing is, I have a nearly identical setup with the Presto MySQL connector, and it runs queries like this instantly. I've checked obvious things like network latency between Presto and SQL Server (it's minimal), but I can't figure out why this is happening. Any tips or debugging steps you can suggest would be a huge help!

内容的提问来源于stack exchange，提问作者Umair Afzal