VLM-MAX Sizing Guidance

Sizing Your VLM-MAX

The following table provides sizing guidance for various Virtual LoadMaster (VLM) profiles defined by Kemp and the associated performance metrics achieved.

Test Methodology

Kemp performed hands-on lab testing of a VLM-MAX instance with different virtual hardware profile configurations using LoadMaster version 7.2.50. The testing was focused on generating sizing guidance to help end users allocate the resources required for their application needs. Virtual services with two-arm configurations were used, which are common in customer deployments.

The hardware hypervisor platform that was chosen was a Dell R730 server with an Intel Xeon E5-2687 3.0 GHz processor, 8 x 10GbE NICs, and 128GB of memory. The hypervisor platform that was chosen was VMware ESXi 6.7.

Testing was completed using an industry standard network testing tool, Ixia with the IxLoad software module to generate the simulated network traffic.

L7 Throughput

To test the VLM-MAX Layer 7 throughput in a large data transfer environment, the Ixia was used to generate as many TCP connections as possible. Each TCP connection generated 100 HTTP requests, and transferred a 512KB file per request, maximizing the amount of data transferred. The default Maximum Transmission Unit (MTU) for an ethernet network of 1500 bytes was used during all tests.

SSL Throughput

To test the VLM-MAX SSL throughput in a large data transfer environment, the Ixia was used to generate as many SSL Transactions per second as possible. Each SSL Transaction consists of opening a TCP connection, establishing an SSL handshake, requesting a payload of 512KB, and then closing the connection, maximizing the amount of data transferred. This test is likely more representative of many applications today where all the data is encrypted and needs to be handled appropriately by the load balancer.

SSL TPS

The SSL transaction per second test evaluated the performance of the VLM-MAX in an environment where the web servers must handle a large number of simultaneous users using encrypted communication. Each user connecting to the web server generates one or more encrypted HTTP requests using SSL.

To simulate this environment, we configured the Ixia to generate a single HTTPS transaction per TCP connection, using TLS 1.2. This configuration generated as many HTTPS requests as possible, and we observed how many SSL transactions VLM-MAX could process per second with this each VLM profile configuration.

Because the VLM-MAX CPU handles all encryption, and encryption places heavy demands on the CPU, the number of SSL requests is directly dependent on the number and power of the CPUs installed in the system. All SSL testing was carried out using 2K keys which is more typical for modern web applications.