In this week’s real-time analytics news: MLCommons announced the public release of the MLPerf Client v0.5 benchmark for evaluating consumer AI performance.
Keeping pace with news and developments in the real-time analytics and AI market can be a daunting task. Fortunately, we have you covered with a summary of the items our staff comes across each week. And if you prefer it in your inbox, sign up here!
MLCommons announced the public release of the MLPerf Client v0.5 benchmark. This benchmark sets a new standard for evaluating consumer AI performance, enabling users and the industry to measure how effectively laptops, desktops, and workstations can run cutting-edge large language models (LLMs).
Key Features of the MLPerf Client v0.5 benchmark:
- AI model: The benchmark’s tests are based on Meta’s Llama 2 7B large language model, optimized for reduced memory and computational requirements via 4-bit integer quantization.
- Tests and metrics: Includes four AI tasks—content generation, creative writing, and text summarization of two different document lengths—evaluated using familiar metrics like time-to-first-token (TTFT) and tokens-per-second (TPS).
- Hardware optimization: Supports hardware-accelerated execution on integrated and discrete GPUs via two distinct paths: ONNX Runtime GenAI and Intel OpenVINO.
- Platform support: This initial release supports Windows 11 on x86-64 systems, with future updates planned for Windows on Arm and macOS.
- Freely accessible: The benchmark is freely downloadable from MLCommons.org, empowering anyone to measure AI performance on supported systems.
IBM unveiled research in optics technology that could dramatically improve how data centers train and run generative AI models. Researchers have pioneered a new process for co-packaged optics (CPO) to enable connectivity within data centers at the speed of light through optics to complement existing short-reach electrical wires. By designing and assembling the first publicly announced successful polymer optical waveguide (PWG) to power this technology, IBM researchers have shown how CPO will redefine the way the computing industry transmits high-bandwidth data between chips, circuit boards, and servers.
Why it matters: Although data centers use fiber optics for their external communications networks, racks in data centers still predominantly run communications on copper-based electrical wires. These wires connect GPU accelerators that may spend more than half of their time idle, waiting for data from other devices in a large, distributed training process, which can incur significant expense and energy. IBM researchers have demonstrated a way to bring optics’ speed and capacity inside data centers.
Other real-time analytics news in brief
Red Hat announced the latest release of Red Hat Enterprise Linux AI (RHEL AI), Red Hat’s foundation model platform for more seamlessly developing, testing, and running generative artificial intelligence (gen AI) models for enterprise applications.
RHEL AI 1.3 extends Red Hat’s commitment to Granite LLMs with support for Granite 3.0 8b English language use cases. Granite 3.0 8b is a converged model that supports not only English but also a dozen other natural languages, code generation, and function calling. Non-English language use cases, as well as code and functions, are available as a developer preview within RHEL AI 1.3, with the expectation that these capabilities will be supported in future RHEL AI releases.
Aerospike unveiled the latest version of Aerospike Vector Search, featuring new indexing and storage innovations that deliver real-time accuracy, scalability, and ease of use for developers. These advancements simplify deployment, reduce operational overhead, and enable enterprise-ready solutions for just-in-time generative artificial intelligence (GenAI) and machine learning (ML) decisions.
Airbyte announced the availability of data connectors for Oracle databases and Workday for high-performance, secure, and reliable movement of data for users of Airbyte Self-Managed Enterprise and Airbyte Teams. Airbyte builds and maintains the connectors to provide enhanced capabilities and support for critical enterprise systems. The Oracle connector also supports Change Data Capture (CDC) – a process that identifies and tracks changes to a database and delivers only those changes for greater efficiency and time savings.
Algolia unveiled Algolia Data Transformations, a new data preparation tool to improve the quality of data to be indexed by customers. Using this new tool, developers can apply Extract, Transform, Load (ETL) functions to enrich the data indexed in Algolia, leading to superior search and retrieval results. The tool simplifies even the most complex data preparation tasks, optimizing data for more precise search outcomes and seamless discovery experiences.
Cloudera announced the launch of its Retrieval-Augmented Generation (RAG) Studio. RAG Studio empowers enterprises to deploy RAG chatbots using their real-time enterprise data in just minutes. This no-code solution makes AI applications more accessible to non-technical users, fosters collaboration between business and IT teams in AI development, and democratizes AI tools for broader user adoption.
In other Cloudera news, the company announced that CrewAI has joined the Cloudera Enterprise AI Ecosystem to revolutionize multi-agentic driven workflows. This strategic collaboration aims to unlock value from enterprise data by enabling intelligent, autonomous processes that can continuously adapt, learn, self-heal, and take action in real time.
Confluent announced the general availability of the Confluent Platform for Apache Flink with added enterprise-level security capabilities and easier ways to manage and scale on-premises Apache Flink workloads. In addition, Confluent announced WarpStream Orbit for easier migration to WarpStream’s “Bring Your Own Cloud (BYOC)” deployment model.
Dataiku announced the launch of Dataiku Stories, a Generative AI-powered data storytelling solution that enables business users to quickly and easily generate insights and transform company data into visual presentations on their own. Dataiku Stories bridges the gap between static slide presentations, which are easy to create and share but often showcase outdated or untrusted data, and enterprise dashboards, which can be complex to build and maintain but access governed data.
EnterpriseDB (EDB) introduced enhancements to EDB Postgres AI, empowering enterprises to deploy secure, flexible AI-driven applications in a sovereign, hybrid environment. EDB Postgres AI delivers a single pane of glass (SPoG), combining cloud agility with a hybrid-first intelligent platform tailored for transactional, analytical, and AI workloads—allowing organizations to accelerate their AI initiatives from development to production-ready applications.
IBM announced that its observability solution, IBM Instana, is now powered with automated resource optimization by IBM Turbonomic and is generally available. The integration of real-time observability capabilities of IBM Instana with the resource optimization capabilities of IBM Turbonomic enables companies to run their critical applications smoothly 24×7 in a cost-effective way, empowering their IT operations to proactively act before problems affect their customers and their baseline.
KNIME announced the launch of its AI companion – K-AI – to all users. With K-AI, users can co-create powerful data workflows with AI. K-AI will answer questions, make recommendations, and extend or build whole data workflows based on user prompts. The AI companion speeds up the time to insight while giving users complete transparency and control over what AI is doing.
Precisely announced the availability of its real-time change data capture capabilities on Google Cloud Marketplace. Google Cloud users can build data pipelines that replicate data from their legacy data systems, including IBM Z, IBM i, and Oracle, to Google Cloud destinations, such as BigQuery. This seamless connection to data helps organizations increase agility, reduce operational costs, and accelerate innovation.
Quest Software announced two updates designed to further accelerate the enterprise adoption of PostgreSQL. Foglight with the Performance Investigator for PostgreSQL add-on, which is a monitoring and optimization platform providing database observability for the most modern data applications; and SharePlex, a data replication and migration offering that now synchronizes data among PostgreSQL 17.0, Oracle 23ai, MariaDB 11.4.2, Google AlloyDB and Google AlloyDB Omni. Users can leverage these solutions to streamline migrations to PostgreSQL and then optimize its performance and availability.
Solace announced the addition of micro-integrations to its event-driven integration and streaming platform, Solace PubSub+ Platform. The new Solace PubSub+ Micro-Integrations are small, lightweight, event-driven integration modules that connect enterprise technologies – including legacy and SaaS applications, messaging services, databases, files, AI agents, etc. – to an event-driven distribution layer, called an event mesh, enabling information exchange in real-time.
Telmai announced significant enhancements to its enterprise data quality platform, introducing automated workflows designed to accelerate AI adoption. The new capabilities enable organizations to automatically monitor, validate, and optimize data quality across their AI implementations while ensuring regulatory compliance and data reliability at scale.
If your company has real-time analytics news, send your announcements to [email protected].
In case you missed it, here are our most recent previous weekly real-time analytics news roundups:
- Real-time Analytics News for the Week Ending December 7
- Real-time Analytics News for the Week Ending November 23
- Real-time Analytics News for the Week Ending November 16
- Real-time Analytics News for the Week Ending November 9
- Real-time Analytics News for the Week Ending November 2
- Real-time Analytics News for the Week Ending October 26
- Real-time Analytics News for the Week Ending October 19
- Real-time Analytics News for the Week Ending October 12
- Real-time Analytics News for the Week Ending October 5
