AWS Launches Aurora DSQL to Eliminate Global Database Trade-offs
Amazon Web Services has introduced Aurora DSQL, a serverless distributed SQL database engineered to deliver high availability, strong consistency, and scalability across multiple regions. Unveiled during AWS re:Invent 2024, Aurora DSQL is designed to address the limitations of existing distributed databases by offering a solution that combines low latency, SQL compatibility, and minimal operational overhead.
Aurora DSQL features an active-active architecture that allows applications to perform read and write operations from any regional endpoint, ensuring strong data consistency and durability. This design supports 99.999% availability across multiple regions and 99.99% within a single region, eliminating the need for manual failover processes and reducing the risk of data inconsistencies during outages.
The database is compatible with PostgreSQL, enabling developers to utilize familiar tools and frameworks. However, it currently supports a subset of PostgreSQL features, with certain functionalities like foreign keys, views, triggers, and sequences not yet available. Aurora DSQL employs optimistic concurrency control to enhance throughput and system efficiency by avoiding resource locking during transaction execution.
ADVERTISEMENT
To achieve strong consistency with low latency across regions, Aurora DSQL decouples transaction processing from storage, allowing parallelized writes and reducing the need for multiple data exchanges. It also integrates the Amazon Time Sync Service, which synchronizes operations globally with microsecond-level accuracy using satellite-connected atomic clocks.
Aurora DSQL automatically scales compute, I/O, and storage resources based on workload demands, removing the need for manual provisioning or infrastructure management. This serverless approach ensures that applications can handle varying workloads without performance degradation or downtime.
Several organizations, including Razorpay, Electronic Arts, Klarna, QRT, and Autodesk, are exploring Aurora DSQL to support their global operations. Razorpay aims to leverage the database's multi-region strong consistency to enhance the resilience and scalability of its financial applications.
Hashtags
Try Our AI Features
Explore what Daily8 AI can do for you:
Comments
No comments yet...
Related Articles
Channel Post MEA
3 days ago
- Channel Post MEA
Dynatrace Unveils Live Debugger To Ease Cloud Debugging
Dynatrace has announced positive customer adoption of the general availability of Dynatrace Live Debugger. First introduced in January 2025, Live Debugger is the first and only live debugging capability that seamlessly debugs thousands of services concurrently in a production environment without interrupting running code and accelerates developer adoption through its intuitive self-service tools. As enterprise cloud and AI application environments become more dynamic and complex, traditional approaches to debugging applications are no longer sufficient to meet these demands. Dynatrace Live Debugger addresses this challenge head-on, delivering real-time, non-intrusive insights that accelerate problem-solving and streamline performance monitoring at scale. Early adopters, including TELUS, have reported reducing debugging times up to 95%, making it indispensable for enterprises navigating the demands of modern innovation. Capable of supporting thousands of always-on instances and developer self-service, Live Debugger meets the requirements of the most complex modern enterprises embracing cloud and AI-native technologies in Amazon Web Services (AWS), Microsoft Azure, Google Cloud Platform (GCP), RedHat OpenShift, and other enterprise cloud environments. Whether debugging cloud-native deployments or AI workloads, Live Debugger provides a scalable, intelligent solution, which reduces the time, cost, and complexity to allow businesses to operate at speed without compromise. Key features of Live Debugger Instant Data Access : Retrieve critical code-level data on demand without adding new code or redeploying. This eliminates delays that can stretch debugging cycles for weeks or even months. Live Debugger accesses production systems through the Dynatrace platform which provides enterprise-level privacy and security controls for debugging. : Retrieve critical code-level data on demand without adding new code or redeploying. This eliminates delays that can stretch debugging cycles for weeks or even months. Live Debugger accesses production systems through the Dynatrace platform which provides enterprise-level privacy and security controls for debugging. Non-Breaking Breakpoints : Inspect an application's full state, including variables, stack traces, and more, without interrupting or breaking its execution. Live Debugging avoids legacy debugging techniques that compromise performance. : Inspect an application's full state, including variables, stack traces, and more, without interrupting or breaking its execution. Live Debugging avoids legacy debugging techniques that compromise performance. Enterprise Scale: Debug across thousands of workload instances simultaneously in real time to easily identify unknown unknowns faster and free time for innovation. With minimal performance impact, the technology can run in the background, allowing for highly flexible and comprehensive coverage. 'Dynatrace Live Debugger delivers game-changing productivity and insights for businesses managing increasingly complex cloud and AI workloads,' said Steve Tack, Chief Product Officer. 'By enabling real-time debugging with non-breaking breakpoints, we provide developers with the data they need to succeed while protecting application performance. We're addressing a tremendous market opportunity, empowering developers with an experience built for the complex agentic AI environments of the future.' 0 0
The National
3 days ago
- The National
The 'data-driven battlefield' powering change in Formula One
In a sport like Formula One, milliseconds make all the difference. Just to underline that, one of the qualifying sessions in the run up to the last Austrian Grand Prix saw less than 0.8 seconds separate the lap times of the fastest 20 cars. It was a record at the time but you can expect that digit to drop significantly in the coming years. This upsurge in competitiveness comes as a direct result of the correct analysis of the choicest chunks gleaned from a massive upsurge in data collection. AI plays its part. In just a few years, this aspect of the sport has become as crucial to F1 's key players as the drivers themselves. Amazon Web Services (AWS) sits centre stage here, having already played a pivotal role in transforming the racing calendar and acting as the sport's digital backbone. The data it collects and shares has impacted everything from car design and race strategy to fan engagement. These days, race cars are fitted with more than 300 sensors, generating around 1.1 million data points per second. Similarly, there are sensors all around the track at strategic points providing more detail (if any were needed). This colossal amount of information is streamed to AWS, where it is processed and analysed in real-time. Thereafter, each team receives all the data about their own two vehicles, but none concerning their rivals. This allows principals to make split-second decisions on everything from strategy and pit stops to energy deployment and tyre selection, all while the race is happening. The whole process is called predictive modelling and, using historical data and current telemetry, AWS's AI machine learning models are able to predict various outcomes, even the best time for a potential overtaking manoeuvre. 'The partnership with AWS enables us to use machine learning and cloud technologies to improve step-by-step in every department," said Ferrari team principal Frederic Vavaseur. "This can be highly beneficial with everything from product improvements to increasing fan engagement.' Crucially, one of the most compelling innovations is the ability of the system to generate graphics that simulate what would have happened if a team had made a different strategic decision – a different tyre compound choice, say – or even assessing the impact of a driver error. Longer term, the information collated can be used to improve car design, with its ability to run thousands of simulations much more quickly than traditional on-track testing would allow – and also at a fraction of the cost. AWS also offers interactive fan experiences like the Real-Time Race Track online tool, which allows fans to design their own F1 circuits, which are then analysed by company's AI systems to produce projected lap times, top speeds, and even viable race stategies. Spectators can also access F1 Insights, a series of real-time on-screen graphics that appear during broadcasts and provide those watching with information about driver performance, car capabilities, and team strategy. Broadcasters are able to get in on the action, too, as the AWS's Track Pulse tool can quickly mine an extensive stream of historical data to pull up facts and statistics to enliven commentary. It's all about what the company calls intelligent storytelling. Ruth Buscombe-Divey is AWS's motorsports ambassador as well as one of F1 TV's most prominent presenters. She says the changes that have happened within the sport recently are game changers for its future. 'If you applied the technologies that were being used to win races five years ago, it's not going to work,' she says, describing what is happening now as a 'data-driven battlefield' where those involved have to keep pushing both technological advances and how exactly to best use the spiralling amount of information being generated. She sees the changes as crucial to the longevity of the sport. Around 750 million people watch F1 over the course of a season, a figure she cites as being a 'great motivation' in keeping up the momentum to make racing more competitive through the greater use of data. Julie Souza leads sports globally for AWS, driving innovation in data not just in F1 but spectator-led content across the board. She says she is often asked if the influx of new information is going to ruin fans' enjoyment of what they're watching. However, she says it's all about spectators watching sport the way they want to watch it and personalising the experience as much as possible. 'It's very easy for people to go, gosh, this is going to make it too heavy for me,' she says with regard to the recent statistical influx. 'However, if we're talking about this data in a way that alienates, we're doing it wrong. 'The whole point is for this information to make people better understand what they're seeing and enhance their appreciation of the exceptional abilities of the people involved.' The next round of the F1 championship takes place on June 1 at the Circuit de Barcelona-Catalunya in Spain.
The National
4 days ago
- The National
Nvidia gives strong forecast despite chip export controls
Nvidia on Wednesday forecast strong growth for the second quarter of the 2026 fiscal year, despite a loss in revenue due to export controls on its H20 AI chips to China. Nvidia expects revenue of $45 billion in the second quarter, plus or minus 2 per cent. That comes after taking an $8 billion hit due to export control limits related to the H20 chip. Nvidia beat first-quarter estimates, reporting revenue of $44.1 billion, up 69 per cent from a year ago. Analysts polled by Factset expected the company to record first-quarter revenue of $43 billion. Nvidia said it made a profit of $18.8 billion. Data centre revenue came in at $39.1 billion, compared to $22.5 billion for the same time last year. The earnings came weeks after Nvidia announced major partnerships with the UAE and Saudi Arabia during US President Donald Trump 's visit to the Gulf. Mr Huang was also present at the unveiling of the UAE-US AI Campus in Abu Dhabi during Mr Trump's visit. NVidia has also joined OpenAI, Oracle and Abu Dhabi's G42 to build Stargate UAE, which will operate in the campus. And before Mr Trump's visit to Riyadh, Nvidia joined Amazon Web Services and Advanced Micro Devices to collaborate with Humain, the Saudi Public Investment Fund's new AI company. Nvidia has a similar agreement in Taiwan. "We have a line of sight to projects requiring tens of gigawatts of Nvidia AI infrastructure in the not-too-distant future," Nvidia chief financial officer Colette Kress said during a conference call. Chief executive Jensen Huang said: 'Countries around the world are recognising AI as essential infrastructure – just like electricity and the internet – and Nvidia stands at the centre of this profound transformation." Mr Huang also said global demand remains "incredibly strong". Meanwhile, the Blackwell NVL72 AI supercomputer, which he called a "thinking machine designed for reasoning", is in full-scale production. Nvidia shares rose more than 5 per cent to $140.09 during after-hours trading. Its stock was virtually flat when trading closed on Wednesday, 0.51 per cent lower at $134.81 a share. Nvidia was among a number of tech giants that suffered large drops in share price after Mr Trump first announced his larger-than-expected tariff policy last month. But like other companies trading on Wall Street, it has since recovered. China drove much of the attention on Wednesday after the Trump administration had placed stricter export controls on Nvidia's H20 AI chip in April. Nvidia said it incurred a $4.5 billion charge in the first quarter of the 2026 fiscal year because of those controls on China. It had previously warned the latest export rules would cause it to take a hit of $5.5 billion in inventory write-off. During the conference call, Mr Huang said Nvidia cannot reduce its Hopper chip further to comply with the latest chip controls. "As a result, we are taking a multibillion-dollar write-off on inventory that cannot be sold or repurposed," Mr Huang said. Commenting on the latest controls, he said shielding Chinese companies from American competition "only strengthens them abroad and weakens America's position". 'Nvidia was banking that the Trump administration was not going to block their exports of these H20 AI chips,' said Gregory Allen, director of the Wadhwani AI Centre at the Centre for Strategic and International Studies 'The Trump administration has gotten tougher on China even as it has gotten lighter on restrictions placed on other places around the world, like the UAE, like Saudi Arabia,' Mr Allen told CNBC earlier on Wednesday.
