benchANT Customer Stories

On behalf of ScyllaDB, we wrote a detailed comparison of MongoDB and ScyllaDB with over 133 performance measurements.

The measurements include

1. 5 different workloads
2. 3 scaling sizes up to 10 TB of data
3. numerous technical "nuggets", such as 12h runs and measurements with consistency and data model variation.

DBtune is an AI-based DBMS tuning expert. Learning this AI requires many datasets on performance and utilization per workload and resources.

The benchANT benchmarking API was made for this use case. Not only could the required data sets be created, but also the effects of the tuning parameters could be tested.

• What performance does your own server hardware and database configuration deliver?
• How competitive is it compared to current cloud infrastructure and DBaaS products?
• Is a modernization of the infrastructure necessary?

This question was asked by one of the leading German e-commerce companies in the automotive sector - RSU GmbH.

On behalf of Transaction GmbH as a small, unknown provider of the relational database Transbase, basic performance measurements were performed

• Performance analysis of different versions
• Sweet spot analysis
• Competitor comparison for different workloads

On behalf of HarperDB as a new NoSQL DBMS vendor, performance measurements were performed with several objectives:

• Performance analysis of different versions
• Sweet spot analysis
• Competitive comparison for different workloads

On behalf of the IT department of SEW Eurodrive, we conducted a workshop lasting several days on the topic of "Big Data Storage Systems". The following questions were answered in detail:

• How are modern distributed systems structured?
• How can modern data storage IT landscapes be designed?
• How is the operation of cloud-native databases?
• Comparison of different Big Data applications & workloads
• How can decisions for DBMS and DBaaS be made sensibly?
• Detailed interactive case study

Which Database-as-a-Service product delivers the best price-performance ratio?

And how do the internal on-premise servers compare in terms of performance?

This fundamental question was evaluated in an IT project with Bosch Central IT in numerous performance measurements in the cloud and on-premise infrastructure.

How can data-focused startups rely on a scaling and cost-efficient infrastructure that fits their problem from the start?

For the wallpaper app "Quotely:

• we analyzed the existing infrastructure and software architecture
• identified alternative technologies
• and calculated the costs for different architecture & technology scenarios.

Why this was important?

It's here!

Which database management system delivers a reliable throughput of 1,000 ops/s even if one or more database nodes fail?

We evaluated this problem with Daimler TSS in our first PoC.

With 90 performance benchmarks to

• a high-performance database,
• in a highly available multi-node setup
• on “right-sized”, private cloud VMs

for the customer-specific IoT issue and the specific time-series workload.