Buildings scale with good architecture.
Same as our database.

Scaling capabilities

From its very beginning DVTDS was designed to exploit parallel hardware as much as possible. Scalability, throughput and response times were always our focus. The result is a set of advanced features not found in other solutions:
  • The internal architecture is multi threaded.
  • Either individual server node distributes the incoming load automatically across the available CPU cores.
  • Multiple network cards can be used in parallel.
  • Multiple hard disks can be used in parallel.
  • On NUMA hardware architectures multiple server nodes can run in parallel with individual CPU affinity on the same machine.
  • Data and indexes can be distributed across an arbitrary number of nodes. The product imposes no limit to scale dynamic throughput and static storage capacity.
  • Inter - node traffic is optimized and compressed to minimize internal data center traffic.


The clients are not aware of the complex architecture behind their access point because DVTDS knows the right node to handle the client request. The message is submitted internally to the target server which performs the operation. There is no limitation in the number of storage nodes. Any node may further federate data to other types of subordinate servers, including products from other vendors.