Very slow query performance Loki 2.4.1

Hi,

We’re experiencing very slow query performance with loki 2.4.1
We are working in a distributed configuration over k8s with 6 queriers and 2 query frontends.
Our stats look like this:

Ingester.TotalReached 		 16
Ingester.TotalChunksMatched 	 0
Ingester.TotalBatches 		 0
Ingester.TotalLinesSent 	 0
Ingester.HeadChunkBytes 	 0 B
Ingester.HeadChunkLines 	 0
Ingester.DecompressedBytes 	 0 B
Ingester.DecompressedLines 	 0
Ingester.CompressedBytes 	 0 B
Ingester.TotalDuplicates 	 0
Store.TotalChunksRef 		 12227
Store.TotalChunksDownloaded 	 50
Store.ChunksDownloadTime 	 2.862452622s
Store.HeadChunkBytes 		 0 B
Store.HeadChunkLines 		 0
Store.DecompressedBytes 	 120 MB
Store.DecompressedLines 	 64611
Store.CompressedBytes 		 21 MB
Store.TotalDuplicates 		 0
Summary.BytesProcessedPerSecond 	 8.8 MB
Summary.LinesProcessedPerSecond 	 4711
Summary.TotalBytesProcessed 		 120 MB
Summary.TotalLinesProcessed 		 64611
Summary.ExecTime 			 13.714739257s

I’ve seen blog posts with throughput of 4GB/s so definitely there’s an issue here.
CPU and memory both look fine.
Configuration is this (removed URLs of course):

auth_enabled: false
chunk_store_config:
  max_look_back_period: 0s

server:
  http_listen_port: 3100
  grpc_server_max_recv_msg_size: 64000000

distributor:
  ring:
    kvstore:
      store: memberlist
memberlist:
  join_members:
    - loki-distributed-memberlist

ingester:
  lifecycler:
    ring:
      kvstore:
        store: memberlist
      replication_factor: 1
  chunk_idle_period: 3m
  chunk_block_size: 262144
  chunk_encoding: snappy
  chunk_retain_period: 1m
  max_transfer_retries: 0
  wal:
    dir: /var/loki/wal

schema_config:
  configs:
  - from: 2022-01-05
    store: aws
    object_store: s3
    schema: v11
    index:
      prefix: prefix_index_
      period: 24h

storage_config:
  aws:
    s3: s3://XXX
    dynamodb:
      dynamodb_url: dynamodb://ZZZ

limits_config:
  unordered_writes: true
  ingestion_rate_mb: 2048
  ingestion_burst_size_mb: 4096
  per_stream_rate_limit: 30MB
  per_stream_rate_limit_burst: 150MB
  max_streams_per_user: 0
  max_global_streams_per_user: 0

table_manager:
  retention_deletes_enabled: false
  retention_period: 0s
  index_tables_provisioning:
    provisioned_read_throughput: 10000

query_range:
  align_queries_with_step: true
  max_retries: 5
  split_queries_by_interval: 15m
  cache_results: true
  results_cache:
    cache:
      enable_fifocache: true
      fifocache:
        max_size_items: 1024
        validity: 24h

frontend_worker:
  frontend_address: YYY:9095

frontend:
  log_queries_longer_than: 5s
  compress_responses: true
  tail_proxy_url: http://XXX:3100

I’m not sure how to approach this or where to start… It seems like chunk retrieval is not the issue, but something around processing is just extremely slow.
Any help is much appriciated
Thanks!