LeydenDb system, an innovative high performance temporal NewSQL system to address the demanding needs of at scale IoT, CDR, and PM storage but minimize infrastructure costs.
The system has been designed to maintain linear performance in excess of ten (10) terabytes of data. Utilizing LeydenDb's capability to handle massive data sets most end-user queries are resolved in a just a few milliseconds.
How does the open source community benefit from your work?
We plan a community future edition in our road map
Is the code that you’re going to run 100% open source? If so, what is the URL or URLs where it is located?
Yes, although it is at the high end LeydenDb was designed to run on low cost commodity hardware, testing on a single high powered machine or node (Optane / Intel Gold)
What performance-focused articles has your project published before?
We are intensely interested in exploring the upper end of both compute (CPU) and storage throughput (I/O) would provide a key benchmark opportunity. LeydenDb has been optimized for both spinning disks and NVMe onlow cost hardware, this opportunity will allow us to explor the upper end. Basedon the system specs a realistic goal is to prove out that 6.6 million rows/sec. (or 23.8 billion rows/hr.) of continuous insertion of highly indexed large records.
Please state your contributions to the open source community and any other relevant initiatives
Contributor to compressed bitmap repositories such as https://sdm.lbl.gov/fastbit/ and https://github.com/RoaringBitmap/CRoaring related to NoSQL and NewSQL databases. At presentations I also share source code and benchmark results for example JavaOne: MIDP Programming for Wireless Devices Jun 8, 2000, JINI Connection Technology-based Applications for Wireless Telephony Jun 7, 2000, Real World Telecommunications Applications Jun 15, 1999
Would you be willing to share your analysis and results publicly?
Yes we will be generating presentations, whitepapers, and would love to share our findings.
Are you interested in testing Intel Optane SSDs with Intel Memory Drive Technology (IMDT)?
Yes this would be extremely interesting to use e.g. using compressed bitmap technology to address a persistent RAM, e.g. IMDT, image as the backing data store eliminating file-system overhead. This is a major reason why we want to participate in this dem/val lab access opportunity.
Name, email, company, job title
Jon Strabala, CTO LeydenDb 2030 Main Street, Suite 1300 Irvine, CA 92614 USA Email: info AT leydendb DOT io
Two (2) benchmark engineers will collaborate on this effort under my direction.
Project Title and brief description
LeydenDb http://www.LeydenDb.io
LeydenDb system, an innovative high performance temporal NewSQL system to address the demanding needs of at scale IoT, CDR, and PM storage but minimize infrastructure costs.
The system has been designed to maintain linear performance in excess of ten (10) terabytes of data. Utilizing LeydenDb's capability to handle massive data sets most end-user queries are resolved in a just a few milliseconds.
How does the open source community benefit from your work?
We plan a community future edition in our road map
Is the code that you’re going to run 100% open source? If so, what is the URL or URLs where it is located?
N/A
Does the infrastructure provided meet your testing needs (see: https://www.acceleratewithoptane.com/access/)?
Yes, although it is at the high end LeydenDb was designed to run on low cost commodity hardware, testing on a single high powered machine or node (Optane / Intel Gold)
What performance-focused articles has your project published before?
Yes, more of a press release for LeydenDb: http://www.LeydenDb.io
We are intensely interested in exploring the upper end of both compute (CPU) and storage throughput (I/O) would provide a key benchmark opportunity. LeydenDb has been optimized for both spinning disks and NVMe onlow cost hardware, this opportunity will allow us to explor the upper end. Basedon the system specs a realistic goal is to prove out that 6.6 million rows/sec. (or 23.8 billion rows/hr.) of continuous insertion of highly indexed large records.
Please state your contributions to the open source community and any other relevant initiatives
Contributor to compressed bitmap repositories such as https://sdm.lbl.gov/fastbit/ and https://github.com/RoaringBitmap/CRoaring related to NoSQL and NewSQL databases. At presentations I also share source code and benchmark results for example JavaOne: MIDP Programming for Wireless Devices Jun 8, 2000, JINI Connection Technology-based Applications for Wireless Telephony Jun 7, 2000, Real World Telecommunications Applications Jun 15, 1999
Would you be willing to share your analysis and results publicly?
Yes we will be generating presentations, whitepapers, and would love to share our findings.
Are you interested in testing Intel Optane SSDs with Intel Memory Drive Technology (IMDT)?
Yes this would be extremely interesting to use e.g. using compressed bitmap technology to address a persistent RAM, e.g. IMDT, image as the backing data store eliminating file-system overhead. This is a major reason why we want to participate in this dem/val lab access opportunity.