MANGO KESAR-C
Powered by DAOSAI ClustersAll-Flash Storage. Built for AI.
From AI training to inference and agentic workloads, KESAR keeps every GPU fed with data. Consistent performance at any scale, on the Ethernet you already run.
MANGO KESAR-C
Powered by DAOSAI ClustersFrom AI training to inference and agentic workloads, KESAR keeps every GPU fed with data. Consistent performance at any scale, on the Ethernet you already run.
KESAR delivers unmatched performance within a remarkably compact footprint. Begin with the perfect configuration for today, and effortlessly expand your infrastructure at your own pace.
Up to 300 GB/s of RoCEv2 bandwidth from a single 2U controller. Scale out linearly by adding controllers.
Pick a MASS preset and auto-configures your storage, so you reach production on day one, with no storage specialist and no manual tuning required at all.
The redundancy across every controller, network path and drive keeps your data always available.
Manage your data, not storage.
KESAR exposes various native interfaces from a single backend, so AI training and HPC simulation share one storage, with no data copies and no app rewrites.
RoCEv2 block · AI inference KV cache
POSIX file system · migrate as-is
Python / Jupyter · data science
Scientific · weather · genomics
HPC parallel I/O · CFD · FEA
MASS (Mango Accelerated Scalable Storage) is the GUI control plane for KESAR. Run the whole system from a single web dashboard, no command line required.
Bring controllers online and provision storage from one place.
Throughput, IOPS, latency and capacity across the system, in real time.
Integrate with LDAP or Google Workspace, with built-in and custom roles for fine-grained access.
Snapshot-based point-in-time recovery, backup and migration, plus non-disruptive rolling upgrades.
Benchmark milestones, engineering deep-dives and company announcements.
Related news and blog posts will appear here as they are published.
Our team is at the ready to create a customized plan for you to optimize and scale your business.