Extreme Search™ Technology
Self-Searching Enterprise Storage.
Distributed low power processing identifies target data in minutes.
Extreme Search™ solutions begin with NPUsearch™ technology.
FAST DATA SEARCH
Lewis Rhodes Lab’s NPUsearch™ is the backbone of Extreme Search™.
Low power, high performance processing is implemented on FPGA local to SSD. Supported by software compiler and drivers, search is accessed by a simple trio of python commands.
In 2020 Lewis Rhodes Labs introduced NPUsearch™ technology in a self-searching storage appliance that completes regular expression searches on all data files in storage in less than 25 minutes, without indexing or ETL required.
Distributed Search
NPUsearch™ is a hardware and software search engine integrated with SSD in 4 TB storage clusters.
All regular expression searches complete in <25 minutes.
All search is local in storage.
Increasing data volume, more complex queries, same time to search.
No indexing required.
SELF-SEARCHING STORAGE
Multiple appliances can be joined to build racks of storage with no increase in search time. The open-source distributed file system supports fully-scalable search on petabytes of storage files. Multiple servers appear to the user as a single file system.
Currently available systems complete all distributed regex searches within either a 25-minute or a 12-minute time window, depending on specific configurations.
Extremely Parallel Search
NPUsearch™ capability integrated with SSD generates self-searching storage drives.
Collected into 60-100 TB servers it provides the capacity to rapidly identify all target-containing files in local storage in minutes.
Same Search Time
Multiple servers can be racked together for up to petabytes of self-searching storage. Search capacity scales with increased storage, scanning the entire contents in under 25 minutes.
Increasing data volume, more complex queries, and growing storage have the same time to search.
Distributed File System
Users experience a single file system regardless of number of storage servers.