Andromeda is integrating the WEKA NeuralMesh platform as a core AI data storage layer for managed GPU clusters.
“Standardising on WEKA NeuralMesh has given our customers hyperscaler-grade consistency with open-market flexibility at the storage and memory layer, where performance is won or lost,” said Wil Moushey, CEO at Andromeda.
Powering AI workloads
The partnership enables Andromeda’s customers to access a dedicated NeuralMesh or a GPU-native WEKA NeuralMesh Axon deployment to achieve consistent storage performance across any cluster running on any hyperscaler or AI cloud.
It aims to provide the following advantages:
- Ready on day one. New NeuralMesh Axon clusters deploy in as little as 10 minutes.
- Rapid iteration. AI teams building on Andromeda-managed clusters have seen environment startup load times drop from three minutes to 30 seconds.
- Performance at scale. Andromeda clusters running on NeuralMesh Axon are achieving sustained throughput of over 400 GB per second per cluster and over 6 million IOPS in production.
- Native resilience. When faults threaten to cascade across an Andromeda cluster, NeuralMesh contains them and keeps the cluster available.
“Andromeda’s first WEKA NeuralMesh Axon deployment took just 10 minutes to stand up from bare metal. That became our blueprint,” said Vishvajit Kher, lead architect at Andromeda. “If a cloud provider doesn’t offer shared storage, we’re no longer blocked. We deploy WEKA NeuralMesh and know it will perform.”
“Andromeda is solving one of the hardest problems in AI infrastructure: making compute from more than 50 different providers feel like one platform,” said Liran Zvibel, co-founder and CEO at WEKA. “WEKA NeuralMesh extends that promise to the data layer, deploying in minutes and feeding training and inference workloads at full speed on any GPU provider’s hardware, anywhere in the world.”
