Search Results for "without code"

...Genv lets you easily control, configure, monitor and enforce the GPU resources that you are using in a GPU machine or cluster. It is intended to ease up the process of GPU allocation for data scientists without code changes.

The Postgres Operator delivers easy-to-run highly available PostgreSQL clusters on Kubernetes (K8s) powered by Patroni. It is configured only through Postgres manifests (CRDs) to ease integration into automated CI/CD pipelines with no access to Kubernetes API directly, promoting infrastructure as code vs manual operations.

...The back-end daemon accepts data from Syslogd2 or its own client and inserts that data (and envelope information) into one or more databases using MySql client library or dbi drivers (also on SourceForge). The client needs no configuration (all input is via command-line). It is intended for hosts to make database entries without needing database client or interface code. Syslogd2 integration allows DBD2 to act as Syslogd2's database back-end for MySql and other databases supported by the dbi drivers. Original purpose was to track Linux cluster job status but other uses exist.

...In each round, features produced by the network are clustered (e.g. k-means), and the cluster IDs become supervision targets in the next epoch, encouraging the model to refine its representation to better separate semantic groups. This alternating “cluster & train” scheme helps the model gradually discover meaningful structure without labels. DeepCluster was one of the early successes in unsupervised visual feature learning, demonstrating that clustering-based reformulation can rival supervised baselines for many downstream tasks. The repository includes code for feature extraction, clustering, training loops, and evaluation benchmarks like linear probes. Because of its simplicity and modular design, DeepCluster has inspired many later methods.

...Its design embraces OTP principles, using supervised processes and fault-tolerant messaging to keep the registry resilient under failures. Developers can co-locate Swarm with their existing supervision trees to scale workers horizontally without writing custom clustering code. Typical use cases include globally unique workers, sharded consumers, and presence-like coordination where node churn is expected. By turning registry and handoff into configuration and callbacks, Swarm reduces the complexity of multi-node deployments.

JPPF makes it easy to parallelize computationally intensive tasks and execute them on a Grid.

A Java servlet filter which moves session storage out of the java application container, allowing to scale easily by adding more servers, without using a cluster nor changing the code. Session data can be stored into a cookie, memcached, a DB ...

Queue is a load-balancing system popular in the 2000s that lets users control their remote jobs in an intuitive, transparent and nearly seamless way. Development versions feature job migration with and without kernel support. GNU Queue continues to be downloaded despite being decommissioned by the FSF in favor of the newer GNU Parallel project. Older versions of the code are archived here on Sourceforge and some later versions on GNU Savannah.

Hedera provides common interfaces for popular group communications packages like JGroups, Appia, and Spread. Hedera allows applications to switch easily between group communications without code changes and with minimal reconfiguration.