Systems | Development | Analytics | API | Testing

When it comes to managing AI projects, the Command Line Interface (CLI) can be a powerful tool. With ClearML, the CLI becomes an essential resource for creating job templates, launching remote for JupyterLab, VS Code, or SSH development environments, and executing code on a remote machine that can better meet resource needs. Specifically designed for AI workloads, ClearML’s CLI offers seamless control and efficiency, empowering users to maximize their AI efforts.

As compute gets increasingly powerful, the fact of the matter is: most AI workloads do not require the entire capacity of a single GPU. Computing power required across the model development lifecycle looks like a normal bell curve – with some compute required for data processing and ingestion, maximum firepower for model training and fine-tuning, and stepped-down requirements for ongoing inference.

We are excited to announce the launch of our AI Infrastructure Control Plane, designed as a universal operating system for AI infrastructure. With this launch, we make it easier for IT teams and DevOps to gain ultimate control over their AI Infrastructure, manage complex environments, maximize compute utilization, and deliver an optimized self-serve experience for their AI Builders.

Written by Noam Wasersprung, Head of Product at ClearML Last month we released the Resource Allocation & Policy Management Center to help teams visualize their compute infrastructure and understand which users have access to what resources. This new feature makes it easy for administrators to visualize their resource policies for enabling workload prioritization across available resources.

It is a fairly recent trend for companies to amass GPU firepower to build their own AI computing infrastructure and support the growing number of compute requests. Many recent AI tools now enable data scientists to work on data, run experiments, and train models seamlessly with the ability to submit their jobs and monitor their progress. However, for many organizations with mature supercomputing capabilities, Slurm has been the scheduling tool of choice for managing computing clusters.

Our early roadmap in 2024 has been largely focused on improving orchestration and compute infrastructure management capabilities. Last month we released a Resource Allocation Policy Management Control Center with a new, streamlined UI to help teams visualize their compute infrastructure and understand which users have access to what resources.

With the explosion of generative AI tools available for providing information, making recommendations, or creating images, LLMs have captured the public imagination. Although we cannot expect an LLM to have all the information we want, or sometimes even include inaccurate information, consumer enthusiasm for using generative AI tools continues to build.

If you’ve been following our news, you know we just announced free fractional GPU capabilities for open source users, enabling multi-tenancy for NVIDIA GPUs and allowing users to optimize their GPU utilization to support multiple AI workloads as part of our open source and free tier offering.

In our latest research, conducted this year with AIIA and FuriosaAI, we wanted to know more about global AI Infrastructure plans, including respondents’: 1) Compute infrastructure growth plans 2) Current scheduling and compute solutions experience, and 3) Model and AI framework use and plans for 2024. Read on to dive into key findings! Download the survey report now →