(1) 408 430 2503      info@dkube.io

DKube Enterprise      Deep Learning Platform For Your Enterprise Cluster

  • Optimize your AI models through continuous experimentation via hyper-parameter tuning
  • Manage the versions of your models and data sets with full lineage and traceability
  • Enable collaboration across data engineers, data scientists, ML engineers with full audit trails
  • Deploy models in production on the same or different cluster

Industrialization of AI

Industrialization of AI

What It can Learn from the Industrialization of Supermarkets.
Relying on Github or similar methods for code and data repos alone with Tensorflow, PyTorch or similar frameworks is not sufficient to scale AI projects and deployments.

Continue Reading


All the features of DKube Lite

Complete MLOps Workflow

Powerful, flexible, & intuitive operation

Supports workflow from data science through deployment

Formal hand-offs between workflow stages

Production Serving

Structured model management workflow

Tight integration between model development and deployment

Model Catalog supports simple, flexible, & extensible operation

Tracking & History

Full lineage of models for reproducibility & audit

Enables tracking of all inputs to understand impact of changes

Identifies where models and datasets are used

Deep Learning
Mobility & Scalability

Migration from on-prem to cloud, and between cloud providers

Efficient scale-up or scale-out

Support for heterogeneous clusters with device access across servers

Authentication & Authorization

Supports both GitHub & LDAP authorization

Access and privileges granted per user

Integrated support for most Git-based repositories, including GitHub & Bitbucket

API Driven

High availability cluster operation

Kubernetes resiliency through redundancy of pods

DKube resiliency through database redundancy


Support for multi-tenancy

Ability to securely partition system into separate groups

Share project, datasets, models and resources

Optimized for Deep Learning
  • Foundation for the One Convergence DKube Deep Learning-as-a-Service solution
  • Secure multi-tenant partitioning of data and resources
  • Plug and play operation with Kubernetes platform
  • On-demand GPU allocation from flexible pool
  • Support for GPUDirect
Deep Learning
Deep Learning
Distributed, Composable Heterogeneous System
  • Resource pools can be dynamically & securely allocated
  • Pools can be in a server, or distributed among different servers
  • Hosts, devices, and servers can be different types
  • APIs enable features to be used from applications
API Driven
API Driven
  • View & configure system
  • User-specified actions & activities
  • Python and ReST compatible
  • APIs handle the underlying complexity of the network architecture
  • Designed in layers - exposing capabilities for variety of customer application types
Deep Learning
Flexible GUI
  • View & manage network through intuitive GUI
  • Network topology, virtual connections, & key statistics provided
  • Real-time statistics & historical graphs
  • APIs available for scripting or integration into applications
Deep Learning
API Driven
Support & Services
  • Support at every phase of your design
  • Software modules based on customer requirements
  • Production-ready turnkey solutions
  • Tuning & optimization for target markets or specific platforms
  • Experts in Networking, Security, PCIe Fabrics, Deep Learning & Cloud Native Infrastructure