Challenges

Log analytics is characterized by large data sizes, infrequent data reading, and requirements for low cost and fast speed. The current log analytic solutions based on Loki or Elasticsearch often face the following challenges:
Log analytics is characterized by large data size, infrequent data reading, and requirements for low cost and fast speed. The current Loki- or Elasticsearch-based log analytic solutions are often challenged by the following problems:
  • Low writing throughput

    A big system can generate up to tens of millions of log messages per second. For such huge data sizes, common solutions struggle to ensure high writing throughput and real-time data availability at the same time.
  • High storage cost

    Log data grows over time and imposes great pressure on storage. Common solutions often come with high storage costs and make it unsustainable to store logs for a long time.
  • Slow query response

    Log analysis requires high performance in keyword searches to enable quick troubleshooting. However, common solutions are inadequate in this term.

The VeloDB solution

Integrate your log system with VeloDB for real-time, high-performance, and low-cost log management and analysis.
  • High writing throughput

    • ·VeloDB has optimized its inverted index for log analysis and can deliver a writing speed 4 times faster than Elasticsearch.
    • ·A group commit mechanism at the server end helps ensure a writing throughput measured in GB/s and data availability within seconds.
  • Low storage cost

    • ·With columnar storage, simplified inverted index, and a high data compression ratio, VeloDB only consumes 1/5 of the storage space that Elasticsearch uses for the same data size.
    • ·By tiered storage, VeloDB can reduce storage costs by 2/3 compared to Elasticsearch. Factoring in the storage space it saves, VeloDB only costs 1/15 of the storage expenses of Elasticsearch.
  • Fast query response

    • ·VeloDB massively reduces data scanning by data partitioning, bucketing, and timestamp filtering.
    • ·With the inverted index, VeloDB can quickly locate the target log and respond to queries within seconds.

Try VeloDB now

VeloDB Cloud
Fully managed, cloud-native, real-time data warehouse service
Start free
VeloDB Enterprise
Self-managed software on premises, on VMs, or K8s
Start free