Connect with us

Technology

Cockroach Labs adds vector search, updates pricing options

Published

on

Cockroach Labs adds vector search, updates pricing options

Cockroach Labs on Thursday unveiled vector search capabilities aimed at enabling customers to access and operationalize unstructured data to train generative AI models and applications.

In addition, the vendor introduced a new tool designed to improve efficiency by reducing query times and optimizing usage as well as new pricing tiers for CockroachDB Cloud. Each of the new features is part of Cockroach Labs’ CockroachDB 24.2 update.

As enterprise interest in generative AI has exploded in popularity over the past two years, vector search has become a common means of discovering the data — much of it unstructured — needed for the retrieval-augmented generation (RAG) pipelines that feed and train generative AI tools. As a result, adding vector search capabilities is important for database vendors such as Cockroach Labs, according to Stephen Catanzano, an analyst at TechTarget’s Enterprise Strategy Group.

“Vector search is a crucial advancement for CockroachDB because it allows users to handle unstructured data,” he said. “By adding vector search, Cockroach enables users to … manage data more intelligently. This is especially important as enterprises increasingly rely on AI and need databases that can handle vectors for enhanced performance and accuracy.”

Advertisement

Based in New York City, Cockroach Labs is a database vendor that provides a cloud-native SQL database platform.

To date, the vendor has raised more than $600 million in funding, including $278 million in January 2021 and $160 million in May 2020. Competitors, meanwhile, include other database specialists such as MongoDB and Yugabyte as well as database offering from tech giants including Amazon DynamoDB and Microsoft SQL Server.

New capabilities

OpenAI’s November 2022 launch of ChatGPT marked a significant advance in large language model (LLM) capabilities.

Since then, many enterprises have made developing generative AI features a priority, combining LLM capabilities with their own proprietary data to develop models and applications that understand their business.

Advertisement

Using such models and applications, enterprises can develop generative AI assistants that enable users of any skill level to use natural language processing to query and analyze data to make informed decisions. In addition, enterprises can program models and applications to take on repetitive tasks that otherwise need to be performed by data engineers and other experts, thus making those experts more efficient.

However, joining the capabilities of LLMs with proprietary data to train generative AI tools is not simple.

Without large volumes of quality data — and even sometimes with it — generative AI tools are prone to AI hallucinations, which are incorrect and sometimes bizarre outputs that can have serious consequences if not caught by humans. To feed models and applications with enough data to reduce the likelihood of hallucinations, unstructured data is needed.

Unstructured data such as text, images and audio files is estimated to make up over 80% of all data. Without some form of structure, however, data is difficult to operationalize. Vectors, which are numerical representations of data automatically assigned by algorithms, give unstructured data the structure it needs to be searched and discovered.

Advertisement

Therefore, to meet the needs of customers wanting to develop generative AI tools, many database specialists and other data management vendors have added vector search and storage capabilities.

For example, Cockroach Labs competitors including MongoDB and Couchbase now offer vector search and storage while tech giants AWS and Oracle have made vector search and storage central to their database strategies.

Now, Cockroach Labs is introducing its own vector search capabilities, adding tools that are critical for any database vendor as enterprise interest in generative AI surges, according to Kevin Petrie, an analyst at BARC U.S.

CockroachDB’s vector search capabilities are enabled by an integration with pgvector, an open source tool for PostgreSQL databases that uses semantic modeling to improve vector searches. Through the integration, Cockroach Labs customers can now perform semantic searches across large vector datasets to discover data relevant to generative AI models and applications such as recommendation engines and AI assistants.

Advertisement

“Given the popularity of GenAI, vector search has become somewhat of a must-have feature among database vendors,” Petrie said.

In a typical RAG workflow, vector search is a way for enterprises to apply generative AI language models to their own proprietary data, he continued. The vector databases find and retrieve unstructured data such as text or imagery, then feed it into pipelines to make generative AI language models less likely to hallucinate.

“Recognizing this opportunity, many database vendors are adding vector search features,” Petrie said.

While some vendors have had vector search capabilities for more than a year — and vendors such as Pinecone specialize in vector databases — Cockroach Labs is only getting started with vector search. So, although recommendation engines and AI assistants are two target use cases, there are others, Petrie added.

Advertisement

“I’ll be interested to see what additional detail they provide in coming announcements about capabilities, target use cases and ideal datasets,” he said.

In addition to the new vector search capabilities, Cockroach Labs unveiled a new pricing structure for the fully managed version of its database; it also offers a self-managed version.

The vendor now offers CockroachDB Cloud at Basic, Standard and Advanced tiers. Previously, the vendor offered only Serverless and Dedicated tiers.

Basic and Advanced essentially replace Serverless and Dedicated while Standard represents a new tier between the two to give customers three fully managed options.

Advertisement

Basic begins at no cost with customers incurring charges once they exceed 10 gigabytes of storage and 50 million request units per month. Standard begins at $146 per month per two vCPUs (virtual CPUs) and Advanced begins at $295 per month per two vCPUs.

Beyond simply renaming two of its pricing options and adding a new one, the new pricing tiers are designed to better match an enterprise’s workload needs to a pricing tier, according to Cockroach Labs CEO Spencer Kimball.

For example, the Basic tier might be best for an organization with entry-level workloads whereas the Advanced tier is the likely fit for an enterprise requiring high security and scalability. Meanwhile, the Standard tier offers a balance that is intended to provide the cost efficiency of Basic with some of the efficiency, scalability and security of Advanced.

“The introduction of the Standard tier [enables] companies to consolidate a range of workloads while optimizing cost and performance,” Kimball said.

Advertisement

Catanzano likewise said the addition of new pricing tiers is significant given that they provide both existing and potential customers with flexibility as their workload demands and budgets change.

“It simplifies cloud adoption and makes CockroachDB accessible to a wider range of users, supporting scalability from startups to large enterprises,” he said.

Beyond new vector search capabilities and reorganized pricing, Cockroach Labs unveiled Generic Query Plans, a tool that reduces query times to make complex queries more efficient and less expensive by using less compute power.

A mix of customer feedback and responding to market trends provided the impetus for adding vector search and other new features, according to Kimball.

Advertisement

Many enterprises are making generative AI a priority. To meet their needs, Cockroach Labs needed to add the vector search capabilities that enable those enterprises to find and operationalize relevant data as well as improve the performance of its database to handle the workloads that AI demands.

“We’ve designed CockroachDB to meet those evolving needs by making sure our database is ready to handle the scale and complexity of these workloads,” Kimball said.

Looking ahead

With CockroachDB 24.2 now available, Cockroach Labs plans to continue adding capabilities to enable customers to run AI and machine learning workloads, according to Kimball.

Included is the recognition that many enterprises are only getting started with AI and machine learning and that both workload size and complexity will increase over time.

Advertisement

“Our goal is to provide businesses with a database that not only meets today’s demands but is future proofed for tomorrow’s challenges, allowing our customers to stay ahead in a rapidly evolving landscape,” he said.

That focus on adding and improving capabilities that enable customers to develop generative AI models and applications is wise, according to Petrie.

With Cockroach Labs only now starting with vector search, it’s important that the vendor demonstrate its commitment to enabling advanced application development.

“I’ll be interested to see how serious Cockroach is about supporting RAG workflows,” Petrie said. “If they are, I would expect more announcements about the benefits of enriching generative AI language model prompts with both vector and relational data.”

Advertisement

Catanzano likewise suggested that Cockroach Labs continue to add support for customers interested in developing generative AI tools. Just as the integration with pgvector is how Cockroach labs is adding vector search, integrations with other vendors could be a means of quickly developing an ecosystem for AI and machine learning.

“To continue its growth, Cockroach Labs could further integrate more AI-driven data management features such as enhanced support for machine learning workloads and more seamless multi-cloud capabilities,” Catanzano said.

Adding new tools for developers and features such as data observability could also benefit Cockroach Labs and help the vendor stand out from its competitors, he continued.

“These steps could help Cockroach Labs solidify its leadership in cloud-native, resilient databases,” Catanzano said.

Advertisement

Eric Avidon is a senior news writer for TechTarget Editorial and a journalist with more than 25 years of experience. He covers analytics and data management.

Source link

Continue Reading
Advertisement
Click to comment

You must be logged in to post a comment Login

Leave a Reply

Technology

Samsung unboxes the all-new Galaxy S24 FE: Video

Published

on

Samsung unboxes the all-new Galaxy S24 FE: Video

The Samsung Galaxy S24 FE was announced yesterday, and the company has just unboxed it. This unboxing was presented in the form of a promo video which surfaced on the company’s official YouTube channel.

The Galaxy S24 FE unboxing video is now live on YouTube

The unboxing video has a duration of around a minute and a half, and it’s embedded at the very end of the article. It not only shows you what sits in the box, but you get to see all the colors of the device. On top of that, some features are highlighted here too.

In addition to the phone itself, and some paperwork, a SIM ejection pin is included, and a charging cable. That’s a USB-C to USB-C charging cable if you were wondering. No, the charger is not included in the box.

The phone comes in Blue, Graphite, Gray, Mint, and Yellow colors, in case you missed the memo yesterday. This video will show you the device from all angles. That way you’ll see its flat sides, which is a change from last year’s model. The device is also larger this time around, as it has a larger display.

Advertisement

Samsung claims this phone offers “stunning low-light portraits”

Samsung says that the Galaxy S24 FE can provide “stunning low-light portraits” thanks to ProVisual Engine. The cameras are highlighted in the video in general, and the same goes for the SoC. This phone is fueled by the Exynos 2400e chip.

A 6.7-inch AMOLED display with a 120Hz refresh rate is used this time around, instead of a 6.4-inch panel. It’s also brighter now, as it has a peak brightness of 1,900 nits. Gorilla Glass Victus+ protects that display, by the way.

The device is also water and dust-resistant, and so on. If you’d like to know more about the Galaxy S24 FE, check out our original announcement. You can also pre-order the device as we speak, if you’re interested.

Source link

Advertisement
Continue Reading

Science & Environment

WTI heads for weekly loss as supplies rise

Published

on

WTI heads for weekly loss as supplies rise


The oil market today doesn't preemptively price in risk, says S&P Global's Dan Yergin

U.S. crude oil on Friday was on pace for its first weekly loss in three weeks, as the prospect of growing oil supplies from Saudi Arabia overshadowed China’s efforts to stimulate its economy.

The U.S. benchmark West Texas Intermediate is down nearly 6% this week, while global benchmark Brent has pulled back nearly 4%. Prices have fallen even as conflict in the Middle East escalates, with Israel and Hezbollah trading blows in Lebanon.

“It is amazing to see that … war doesn’t affect the price, and that’s because there’s been no disruption,” Dan Yergin, vice chairman of S&P Global, told CNBC’s “Squawk Box” Friday.

Advertisement

“There’s still over 5 million barrels a day of shut in capacity in the Middle East,” Yergin said.

Here are Friday’s energy prices:

  • West Texas Intermediate November contract: $67.51 per barrel, down 16 cents, or 0.24%. Year to date, U.S. crude oil is down more than 5%.
  • Brent November contract: $71.37 per barrel, off 23 cents, or 0.32%. Year to date, the global benchmark is down about 7%.
  • RBOB Gasoline October contract: $1.9596 per gallon, little changed. Year to date, gasoline is down about 7%.
  • Natural Gas November contract: $2.774 per thousand cubic feet, up 0.76%. Year to date, gas is up about 10%.

Oil sold off Thursday on a report that Saudi Arabia is committed to increasing production later this year, even if it results in lower prices for a pro-longed period.

OPEC+ recently postponed planned output hikes from October to December, but analysts have speculated that the group might delay the hikes again because oil prices are so low.

The oil selloff erased gains from earlier in the week after China unveiled a new round of economic stimulus measures. Soft demand in China has been weighing on the oil market for months.

Advertisement

“The thing that’s dominated the market is the weakness in China. Half the growth in world oil demand over a number of years has simply been in China, and it hasn’t been happening,” Yergin said.

“The big question is, stimulus, will you see a recovery in China,” he said. “That’s what the market is struggling with.”

Don’t miss these energy insights from CNBC PRO:



Source link

Advertisement
Continue Reading

Servers computers

42U Adjustable Depth Open Frame 4 Post Server Rack Cabinet – 4POSTRACK42 | StarTech.com

Published

on

42U Adjustable Depth Open Frame 4 Post Server Rack Cabinet - 4POSTRACK42 | StarTech.com



The 4POSTRACK42 42U Server Rack lets you store your servers, network and telecommunications equipment in a sturdy, adjustable depth open-frame rack.

Designed with ease of use in mind, this 42U rack offers easy-to-read markings for both rack units (U) and depth, with a wide range of mounting depth adjustments (22 – 40in) that make it easy to adapt the rack to fit your equipment.

This durable 4-post rack supports a static loading capacity of up to 1320lbs (600kg), and offers compliance with several industry rack standards (EIA/ECA-310, IEC 60297, DIN 41494) for a universal design that’s compatible with most rack equipment.

For a complete rack solution that saves you time and hassle, the rack includes optional accessories such as casters, leveling feet and cable management hooks. The base is also pre-drilled for securing the rack to the floor if needed, providing you with many options to customize the rack to fit your environment.

Advertisement

Backed by a StarTech.com 2-year warranty and free lifetime technical support.

To learn more visit StarTech.com

source

Continue Reading

Technology

Intel reportedly rebuffed an offer from ARM to buy its product unit

Published

on

Intel reportedly rebuffed an offer from ARM to buy its product unit

Intel’s fortunes have declined so rapidly over the past year that chip designer ARM made a “high level inquiry” about buying its crown jewel product unit, Bloomberg reported. However, Intel said the division wasn’t for sale and turned down the offer, according to an unnamed insider.

There are two main units inside Intel, the product group that sells PC, server and networking chips and a chip manufacturing foundry. ARM had no interest in Intel’s foundry division, according to Bloomberg‘s sources. ARM and Intel representatives declined to comment.

Intel’s fortunes have been on the wane for years, but the decline over the last 12 months has been especially dramatic. Following a net $1.6 billion loss in Q2 2024, the company announced that it was laying off 15,000 employees as part of a $10 billion cost reduction plan. Last week, the company also revealed plans to transform its ailing foundry business into an independent subsidiary. Intel lost half its market value last year and is now worth $102.3 billion.

ARM sells its processor designs to Qualcomm, Apple and other manufacturers (mostly for mobile phones) but doesn’t build any chips itself. Purchasing Intel’s product division would completely transform its business model, though that scenario seems highly improbable.

Advertisement

With Intel wounded at the moment, rivals have been circling. Qualcomm also expressed interest in taking over Intel recently, according to a report from last week. Any mergers related to ARM and Qualcomm would be regulatory nightmares, but the fact that the offers exist at all shows Intel’s vulnerability.

Intel has other avenues to boost investment. Apollo Global Management (the owner of Yahoo and Engadget) has offered to invest as much as $5 billion in the company, according to a recent Bloomberg report. Intel also plans to sell part of its stake in chip-maker Altera to private equity investors.

Source link

Advertisement
Continue Reading

Servers computers

Wallmount Rack Server 9U/Rak Server Ukuran 9U Single Glass Door #servers #komputer

Published

on

Wallmount Rack Server 9U/Rak Server Ukuran 9U Single Glass Door #servers  #komputer



Wallmount Rack Server 9U/Rak Server Ukuran 9U Single Glass Door Di rakit oleh siswa dan siswi jurusan teknik komputer jarngan .

source

Continue Reading

Technology

From cost center to competitive edge: The strategic value of custom AI Infrastructure

Published

on

From cost center to competitive edge: The strategic value of custom AI Infrastructure

Join our daily and weekly newsletters for the latest updates and exclusive content on industry-leading AI coverage. Learn More


This article is part of a VB Special Issue called “Fit for Purpose: Tailoring AI Infrastructure.” Catch all the other stories here.

AI is no longer just a buzzword — it’s a business imperative. As enterprises across industries continue to adopt AI, the conversation around AI infrastructure has evolved dramatically. Once viewed as a necessary but costly investment, custom AI infrastructure is now seen as a strategic asset that can provide a critical competitive edge.

Mike Gualtieri, vice president and principal analyst at Forrester, emphasizes the strategic importance of AI infrastructure. “Enterprises must invest in an enterprise AI/ML platform from a vendor that at least keeps pace with, and ideally pushes the envelope of, enterprise AI technology,” Gualtieri said. “The technology must also serve a reimagined enterprise operating in a world of abundant intelligence.” This perspective underscores the shift from viewing AI as a peripheral experiment to recognizing it as a core component of future business strategy.

Advertisement

The infrastructure revolution

The AI revolution has been fueled by breakthroughs in AI models and applications, but those innovations have also created new challenges. Today’s AI workloads, especially around training and inference for large language models (LLMs), require unprecedented levels of computing power. This is where custom AI infrastructure comes into play.

>>Don’t miss our special issue: Fit for Purpose: Tailoring AI Infrastructure.<<

“AI infrastructure is not one-size-fits-all,” says Gualtieri. “There are three key workloads: data preparation, model training and inference.” Each of these tasks has different infrastructure requirements, and getting it wrong can be costly, according to Gualtieri. For example, while data preparation often relies on traditional computing resources, training massive AI models like GPT-4o or LLaMA 3.1 necessitates specialized chips such as Nvidia’s GPUs, Amazon’s Trainium or Google’s TPUs.

Nvidia, in particular, has taken the lead in AI infrastructure, thanks to its GPU dominance. “Nvidia’s success wasn’t planned, but it was well-earned,” Gualtieri explains. “They were in the right place at the right time, and once they saw the potential of GPUs for AI, they doubled down.” However, Gualtieri believes that competition is on the horizon, with companies like Intel and AMD looking to close the gap.

Advertisement

The cost of the cloud

Cloud computing has been a key enabler of AI, but as workloads scale, the costs associated with cloud services have become a point of concern for enterprises. According to Gualtieri, cloud services are ideal for “bursting workloads” — short-term, high-intensity tasks. However, for enterprises running AI models 24/7, the pay-as-you-go cloud model can become prohibitively expensive.

“Some enterprises are realizing they need a hybrid approach,” Gualtieri said. “They might use the cloud for certain tasks but invest in on-premises infrastructure for others. It’s about balancing flexibility and cost-efficiency.”

This sentiment was echoed by Ankur Mehrotra, general manager of Amazon SageMaker at AWS. In a recent interview, Mehrotra noted that AWS customers are increasingly looking for solutions that combine the flexibility of the cloud with the control and cost-efficiency of on-premise infrastructure. “What we’re hearing from our customers is that they want purpose-built capabilities for AI at scale,” Mehrotra explains. “Price performance is critical, and you can’t optimize for it with generic solutions.”

To meet these demands, AWS has been enhancing its SageMaker service, which offers managed AI infrastructure and integration with popular open-source tools like Kubernetes and PyTorch. “We want to give customers the best of both worlds,” says Mehrotra. “They get the flexibility and scalability of Kubernetes, but with the performance and resilience of our managed infrastructure.”

Advertisement

The role of open source

Open-source tools like PyTorch and TensorFlow have become foundational to AI development, and their role in building custom AI infrastructure cannot be overlooked. Mehrotra underscores the importance of supporting these frameworks while providing the underlying infrastructure needed to scale. “Open-source tools are table stakes,” he says. “But if you just give customers the framework without managing the infrastructure, it leads to a lot of undifferentiated heavy lifting.”

AWS’s strategy is to provide a customizable infrastructure that works seamlessly with open-source frameworks while minimizing the operational burden on customers. “We don’t want our customers spending time on managing infrastructure. We want them focused on building models,” says Mehrotra.

Gualtieri agrees, adding that while open-source frameworks are critical, they must be backed by robust infrastructure. “The open-source community has done amazing things for AI, but at the end of the day, you need hardware that can handle the scale and complexity of modern AI workloads,” he says.

The future of AI infrastructure

As enterprises continue to navigate the AI landscape, the demand for scalable, efficient and custom AI infrastructure will only grow. This is especially true as artificial general intelligence (AGI) — or agentic AI — becomes a reality. “AGI will fundamentally change the game,” Gualtieri said. “It’s not just about training models and making predictions anymore. Agentic AI will control entire processes, and that will require a lot more infrastructure.”

Advertisement

Mehrotra also sees the future of AI infrastructure evolving rapidly. “The pace of innovation in AI is staggering,” he says. “We’re seeing the emergence of industry-specific models, like BloombergGPT for financial services. As these niche models become more common, the need for custom infrastructure will grow.”

AWS, Nvidia and other major players are racing to meet this demand by offering more customizable solutions. But as Gualtieri points out, it’s not just about the technology. “It’s also about partnerships,” he says. “Enterprises can’t do this alone. They need to work closely with vendors to ensure their infrastructure is optimized for their specific needs.”

Custom AI infrastructure is no longer just a cost center — it’s a strategic investment that can provide a significant competitive edge. As enterprises scale their AI ambitions, they must carefully consider their infrastructure choices to ensure they are not only meeting today’s demands but also preparing for the future. Whether through cloud, on-premises, or hybrid solutions, the right infrastructure can make all the difference in turning AI from an experiment into a business driver


Source link
Continue Reading

Trending

Copyright © 2024 WordupNews.com