Kubeflow brings Kubernetes to machine learning workloads

Now in beta, the open source Kubeflow project aims to help deploy a machine learning stack on the Kubernetes container orchestration system.

The Kubeflow machine learning toolkit project is intended to help deploy machine learning workloads across multiple nodes but where breaking up and distributing a workload can add computational overhead and complexity. Kubernetes itself is tasked with making it easier to manage distributed workloads, while Kubeflow centers on making the running of these workloads portable, scalable, and simple. Scripts and configuration files are part of the project. Users can customize their configuration and run scripts to deploy containers to a chosen environment.

To help management deployments, Kubeflow works with Version 0.11.0 or later of the Ksonnet framework, for writing and deploying Kubernetes configurations to clusters. Kubernetes 1.8 or later is required, in a cluster configuration. Kubeflow also works with the following technologies:

  • TensorFlow machine learning models, which can be trained for use on premises or in the cloud.
  • Jupyter notebooks, to manage TensorFlow training jobs.
  • Seldon Core, a platform for deploying machine learning models on Kubernetes.

Kubeflow extends the Kubernetes API by adding custom resource definitions to a cluster, so Kubernetes can treat machine learning workloads as first-class citizens. Described by the open source project as being cloud-native, Kubeflow also integrates with the Ambassador for Ingress and Pachyderm projects for management of data science pipelines. Plans call for extending Kubeflow beyond TensorFlow, with backing considered for the PyTorch and MXNet deep learning frameworks.

Top web browsers 2018: Chrome edges toward supermajority share

Google’s Chrome last month continued to creep up on a two-thirds supermajority of browser share, while Microsoft’s once-dominant position deteriorated. Again.

According to analytics company Net Applications, Chrome’s user share climbed half a percentage point in August, reaching 65.2%, an all-time high. In the last 12 months, Chrome has gained 5.9 percentage points, the only browser of the top four – others include Apple’s Safari, Microsoft’s Edge and Internet Explorer (IE), and Mozilla’s Firefox – to add to its total during that period.

Net Applications calculates user share by detecting the agent strings of the browsers people use to visit its clients’ websites. The firm then tallies the visitor sessions – which are effectively visits to the site, with multiple sessions possible daily – rather than count only users, as it once did. Net Applications primarily measures activity, although it does so differently than rival sources, which total page views.

If the trend of the last 12 months continue, Chrome will take the two-thirds prize in November. Barring any change in the browser battle, Chrome will account for 70% of the global share by June 2019.

The only other browsers to have accumulated that much share since the web broke out of its academia-government ghetto in the 1990s were Netscape’s Navigator and Microsoft’s IE. The former faded under assault from the latter, vanishing for good in early 2008; IE is following in its one-time rival’s footsteps.