How has the recent turmoil within the OpenAI offices changed your plans to use GPT in a business process or product in 2024?
Increased uncertainty means we are more likely to evaluate alternative AI chatbots and LLMs.
No change in plans, though we will keep an eye on the situation.
With Sam Altman back in charge, we are more likely to go all-in with GPT and LLMs.
What recent turmoil?

Latest Aerospike Update Supports Large-Scale Data Models

Aerospike Database 6 promises faster performance with unlimited scale, plus updates including expanded batch processing and document database functionality.
Apr 27th, 2022 12:00pm by
Featued image for: Latest Aerospike Update Supports Large-Scale Data Models

Many businesses need to process data in real time — operations like digital payments, fraud protection, personalized customer recommendations and risk analysis need to happen without delay to be effective and to meet customer expectations.

The world has seen a massive explosion in the amount of data being produced. Global data creation is expected to exceed 180 zettabytes by 2025, up from about 64 zettabytes in 2020, according to market research provider Statista Research.

And, 30% of data generated will be real time by 2025, according to market research firm IDC, the rise fueled by increasing customer demand for access to a product or service whenever and wherever they want, and on any device.

Companies need to use that data with speed and efficiency to prevent costly delays and to take advantage of time-critical opportunities.

“Latency expectations and requirements are constantly headed in one direction: faster,” said Doug Henchen, analyst at Constellation Research.

Businesses that want to leverage real-time data often have specific requirements from the data platform they use.

“Some customers look for best-of-breed data platforms that support a specific need — like fast transactions, high-scale analytical performance, JSON document handling, geospatial analysis or, in this case, super-low-latency real-time performance,” said Henschen.

According to Aerospike, a data-platform company that serves customers including Airtel, Adobe, Experian, Nielsen and PayPal, organizations have to work and compete in the right-now economy, and data drives that economy.

With the release Wednesday of Aerospike Database 6, the company said it delivers submillisecond performance with unlimited scale, along with updates that include batch processing for all operations and improved document database functionality.

“Our team is looking forward to using the new batch and document storage features of the Aerospike Database 6.0,” said Amit Raj, managing consultant at IT services and consulting company Hoonartek.

The new features, combined with Aerospike’s capabilities for data ingestion, data consistency, query, and XDR replication, will promote further business innovation, he added.

Addressing the ‘Right Now’ Problem

The updates offer large companies a tool to help them compete more effectively, according to Holger Mueller, vice president and principal analyst at Constellation Research.

“There are two types of companies in today’s digitally transforming world: the ones who understand how to win in the moments that matter and those who don’t,” said Mueller. “The ones who do will realize that Aerospike Database 6 is the real-time engine they need to effectively compete with larger, complex data sets we’re now seeing in the enterprise.”

Customers often want immediate results and if a business can’t provide them, they’ll turn elsewhere.

“Not every enterprise needs the submillisecond speeds that Aerospike promises at any scale, but it is a need for companies supporting fraud detection, real-time commerce and personalization, digital payments and other super-low-latency applications,” said Constellation’s Henschen.

JSON is a ubiquitous format for data exchange that’s text based, human readable and easily editable. Aerospike Database 6 delivers native support for JSON document models so business applications can evolve quickly and enterprises can remain competitive through expanding the amount of data they use.

“With the addition of support for JSON, Aerospike has widened the types of data and breadth of use cases it can support,” said Henschen.

The latest upgrades move the company toward providing more of a universal database for real-time performance, he added.

Diagram showing how the Aerospike Real-Time Data Platform Works

Scaling Workloads, Reducing the Number of Servers

Traditional data platforms are failing to meet business requirements that demand a combination of real-time data, performance, scale, security and integrated data, according to a report by the market research firm Forrester, commissioned by Aerospike.

With the Aerospike data platform, companies can reduce their total number of servers even as they scale and increase workloads, according to the report.

Relational databases have been around for decades and are widely implemented, but they have scalability issues and require expensive storage systems. If a customer needs to scale, they have to buy more costly proprietary hardware with additional processing power, storage and memory.

Prior to using the Aerospike platform, Forrrester found, additional servers were continually added to legacy architectures to meet performance requirements. But with Aerospike, the overall number of required servers was reduced by 50% to 70% in the first year, with server reductions increased up to 80% by year three as more workloads were powered by Aerospike.

Organizations interviewed for the Forrester study said they experienced better performance despite cutting infrastructure costs, allowing them to invest more in Aerospike’s technology and add more data to existing use cases.

Legacy NoSQL systems are more scalable and flexible than relational systems, but they’re newer and can be less stable, with fewer functionalities than relational databases. They also have compatibility issues with SQL instructions, which means support for work query issues is more complicated and can slow things down.

With Database 6, Aerospike now supports SQL as third-party, plug-and-play integrations via Spark and Presto/Trino connectors.

How Aerospike’s Real-Time Data Platform Works

Aerospike uses a memory architecture that treats SSDs like DRAM. This means the database’s index is put in RAM, where it can be accessed quickly, and data for the database is put in SSDs, which are treated like RAM and can also be rapidly accessed.

Aerospike is a multicloud platform —  it can be used with Microsoft Azure, Amazon Web Services and Google Cloud for example — and it’s a multimodel petabyte-scale database that supports key-value, a document API, graph, and time-series use cases and data models.

The Aerospike platform also supports cross-data-center replication: The contents of one cluster can be replicated to another cluster, so two copies of the same data can be easily available and not limited by geography.

What’s New in Aerospike Database 6

Aerospike has added new features in response to customer demand to not have to use multiple platforms for different use cases.

New updates offered by Database 6 allow existing customers to expand the way they use the Aerospike platform and add value for customers migrating from other databases:

  • Secondary index query capability: Database 6 conducts complex, efficient queries across massive datasets at the speed of primary index access and supports Spark and Presto/Trino connectors with horizontal and vertical scalability to thousands of workers.
  • Large-scale batch processing: Writes to the database can be batched into groups to make them more efficient, so they take less overhead going over the network. Batch processing works for reads, writes, updates and deletes, is more efficient with improved scaling, and works for all workloads, such as read-heavy, write-heavy, update-heavy and mixed.
  • JSON document storage: JSON documents are stored in binary format. Binary representations have better space efficiency, act on data with less processing and offer flexibility by allowing data types that aren’t part of JSON.
  • Document query: Supports querying document data with JSONpath, a query language for JSON, and enables the user to select and extract data from a JSON document.
  • Document indexing: Database 6 now supports indexing at top-level fields and will support nested elements in the next release, the company said.
  • Database 6 is FIPS 140-2 compliant: Federal agencies can securely deploy multiple applications on a large scale.

“All these features are allowing people to do things at a continued scale with very high performance and to be able to use programming models that they’re used to,” said Srini V. Srinivasan, founder and chief technology officer at Aerospike.

“It’s everything we’ve been known for, but we’ve expanded that capability to allow customers and prospects to be able to take advantage of a broader set of use cases and data models.”

The virtual Aerospace Summit 2022 is slated for May 4. You can register here.

Group Created with Sketch.
TNS owner Insight Partners is an investor in: Pragma, Real.
THE NEW STACK UPDATE A newsletter digest of the week’s most important stories & analyses.