Faster processing and increased throughput.
Built to scale.
Benefit from the latest HPC developments without the need to build and maintain a cluster yourself. Through a combination of accelerated computing and the use of large-scale distributed computer clusters, the Bluebee platform guarantees unparalleled throughput and speed. With Bluebee it is perfectly feasible to continue using gold-standard algorithms, while still achieving your throughput objectives.
Acceleration and distribution
In Bluebee’s compute centers, high-performance processing of NGS data is obtained through the combination of acceleration and distribution. Bluebee accelerated several pipeline components through implementation of standardized and generally accepted algorithms on specific co-processors (FPGA, GPU, CPU). These implementations allow for substantial speedup of specific parts of the pipelines. Moreover, Bluebee has built several distributed implementations of pipeline algorithms. Both techniques combined enable large scale, distributed, accelerated and elastic processing of vast sets of genome data.
Automated provisioning of compute and storage capacity
The provisioning of the right infrastructure in its compute centers is completely handled by Bluebee in an automated way. Users are not required to define nodes, cores, memory, or any other infrastructural parameter. Based on the datasets and the pipelines configured, the application will automatically configure and provision the appropriate compute clusters. Users only need to indicate the turn-around time they wish to see applied to their pipeline runs.
It is Bluebee’s belief that the need for faster and cheaper analytics should not be addressed by using short-cut algorithms. Good science remains essential. With Bluebee it is perfectly feasible to continue using gold-standard algorithms such as GATK or BWA, to obtain high sensitivity and specificity while still achieving unparalleled throughput objectives.
Dr. Barbara Hutter Team Leader Clinical Bioinformatics at DKFZ
“Depending on the data, we are typically seeing between 10-20x acceleration for BWA-ALN and 3-4x acceleration for BWA-MEM, which is substantial. Our entire bioinformatics workflow is reduced by 50%, which directly translates in doubling our capacity.”