Analyst(s): Ron Westfall
Publication Date: January 20, 2025
Oracle unveils next-generation intelligent data architecture built to deliver up to 55% faster AI Vector searches, 2.2X faster analytics scan throughput, and 25% faster transaction processing all while maintaining the same price as the previous generation.
What is Covered in this Article:
- Oracle Exadata X11M provides performance boosts at the same price, accelerating all AI, OLTP, and analytics workloads.
- Oracle Exadata X11M spurs efficiency, saves money, power, and time by supporting more workloads, more data, and more users on fewer systems, costing less.
- Exadata X11M is available on-premises, as Exadata Database Machine, Exadata Cloud@Customer, in Oracle Cloud (OCI), and in multicloud environments within Azure, Google Cloud, and AWS.
The News: Oracle introduced Oracle Exadata X11M, the latest generation of the Oracle Exadata platform. Starting at the same price as the previous generation, Exadata X11M is designed to deliver significant performance improvements across AI, analytics, and online transaction processing (OLTP).
Oracle Exadata X11M: Unleashing an Unparalleled Data Intelligent Architecture
Analyst Take: Oracle kicked off 2025 with the unveiling of Exadata X11M, combining intelligent power management with the ability to run mission-critical workloads faster and on fewer systems, helping customers achieve their energy efficiency and sustainability goals. In addition, the same Exadata capabilities are available across public cloud, multi-cloud, and on-premises environments, giving customers the flexibility to deploy and run their Oracle Database workloads wherever they need without any application changes.
The new offering builds on Oracle’s extensive Exadata pedigree and momentum. In the past 12 months, the Exadata portfolio attained innovation milestones such as introducing mission-critical AI at any scale in May 2024 with Exadata System Software 24ai, launching Exadata Exascale in July 2024, and announcing their multi-cloud alliances with Google in June 2024 and AWS in September 2024, in addition to the previous partnership with Microsoft.
Mission-critical AI at any scale capabilities, for instance, ensures Oracle’s AI vector search can be transparently offloaded to intelligent Exadata storage for up to 30x faster AI Vector search queries with automatic parallelization across the storage servers and each storage server independently computing its top-K matches.
Exadata Database Service on Exascale Infrastructure is a new OCI service that makes the power of Exadata available to smaller organizations and projects by running Oracle Database on pools of virtualized compute and storage, dramatically lowering costs.
And Oracle’s multicloud partnerships are reshaping how organizations look at storing and using business data in the cloud. By running OCI Oracle Database services on Exadata in every major hyperscaler cloud, extreme database performance and availability are available as native services in an ever-expanding universe of cloud regions. The co-location of applications and Exadata enables performance breakthroughs taking advantage of complete Oracle Database and Exadata functionality and compatibility.
The key to a successful AI strategy is to first get an organization’s business data in order. Oracle Database 23ai enables organizations to do just that, regardless of data type, so that AI training and inferencing can be highly productive. Without organized business data in a modern database, AI projects will ultimately fail.
In my view, Exadata X11M serves as the next-generation platform that can run Oracle Database 23ai for enterprise-class AI workflows with vector processing taken to the extremes and helps organizations avoid the struggles of data duplication and data consistency between multiple single-purpose databases. Indeed, it is the Enterprise AI Machine for organizations that are serious about using their private business data for real-world AI workloads.
Considering that most enterprise business data reside on Oracle Databases, the company’s strategy to continue innovating with Exadata and Oracle Database 23ai at a time when others have remained stagnant in both database platform architecture and actual databases themselves positions Oracle for minds share and potential market share gains with its multicloud strategy bringing these innovations to more customers than ever before in the cloud era.
Oracle Exadata X11M: Delivering Significant Improvements over Exadata X10M
Exadata X11M is now an even-faster platform due to its implementation on scale-out 2-socket database servers that leverage the latest AMD EPYC 96-core processors. This provides up to 25% faster performance than the previous generation Exadata X10M alongside memory improvements that are up to 33% faster than X10M.
As such, the Exadata X11M platform delivers comprehensive performance enhancements across multiple data-driven processing domains, including accelerated AI vector searches, improved transaction processing speeds with reduced latency, and significantly faster data analytics scanning and query throughput.
Exadata X11M achieves gains in analytics throughput by leveraging its capability to scan data simultaneously in flash storage and Exadata RDMA Memory (XRMEM). This dual-scanning approach enables the platform to process vast amounts of data at unprecedented speeds, significantly enhancing analytical performance. These enhancements are available in all deployment models at the same price as previous systems, enabling customers to reduce costs by doing more work on the same size platform.
In sum, Exadata X11M performance improvements compared to X10M platform include:
- AI Vector Search: Up to 55% faster persistent vector index (IVF) searches with transparent offloading to intelligent Exadata storage. In addition, in-memory vector index queries (HNSW) are up to 43% faster. New software optimizations available across all Exadata platforms provide an even faster AI search with 4.7x more data filtering in storage servers and 32x faster queries when searching binary vectors.
- Analytics: Up to 25% faster analytic query processing, up to100GB/sec for analytic scans from flash on storage servers – 2.2x faster than X10M, and the ability to scan data from XRMEM in storage servers at an incredible 500GB/sec. When combined with other Exadata innovations, this enables customers to run real-time analytics on transactional data and more rapidly scan terabyte to petabyte data warehouses to gain critical business insights.
- OLTP: Up to 25% faster serial transaction processing, up to 25% greater concurrent transaction throughput, and up to 21% lower SQL 8K I/O read latency—now only 14us. This enables customers to do more and spend less by running more transactions faster with a smaller database system footprint.
Exadata X11M: Lapping the Competition Again
By using the latest AMD EPYC processors, Exadata X11M can dramatically outperform competing database systems using generic X86 commodity servers, delivering unprecedented computational capabilities. In my view, Oracle’s consistent ability to deliver decisive competitive advantages across key criteria and metrics provide the warrant for customers and organizations to give top consideration of Exadata X11M. For instance, Oracle can assert that Exadata X11M leads in OLTP latency for cloud with 8 KB I/O latency as low as 14 µs, representing a 70x advantage over Amazon RDS and Azure SQL each at 1000 µs. As a result, running Oracle Database in multicloud environments with Exadata X11M is faster than AWS services on the AWS cloud and likewise faster than Azure services on the Azure cloud.
Exadata X11M delivers clear-cut competitive advantages in OLTP performance on-premises delivering up to 33x (25.2M) more 8KB SQL Read IOPS compared to Pure Storage FlashArray//XL (0.76M) and up to 3.3x more 8 KB SQL Read IOPS vs. Dell PowerMax 8500 (7.5M). These Oracle-sourced estimates are based on single-rack capabilities that also demonstrate scaling advantages with Exadata X11M scaling up to 14 racks in comparison to PowerMax scaling up to only 2 racks.
Moreover, based on database analytics I/O throughput per server, Exadata leads in analytics throughput across the cloud with Exadata X11M multicloud storage servers providing up to a remarkable 60x higher throughput at 500 GB/s versus AWS Redshift (Ra3.16xlarge) at 16 GB/s and up to 30X higher throughput than AWS RDS. As with OLTP, this makes the performance of Exadata X11M for analytics in multicloud environments faster than AWS services on the AWS cloud itself.
Likewise, Exadata leaves the competition in the dust across analytics throughput on-premises comparisons with Exadata X11M, using only 14 storage servers, delivering 300x higher throughput (7000 GB/x) versus Pure Storage FlashArray//XL (22 GB/s) and up to 40x higher throughput versus Dell PowerMax 8500 (175 GB/s). The performance facts and metrics speak for themselves.
Exadata X11M: Unleashing Extreme Efficiency
Plus, Exadata X11M performance strides enable customers to run their workloads on fewer systems, resulting in hardware, cooling, power, and data center space savings. As such, through Exadata utilization efficiencies gains, customers can spend less and save energy by consolidating more systems due to the following consideration:
- More Workloads: Uniquely prioritize latency-sensitive operations such as locks, cache fusion, commits, and OLTP I/O.
- More Data: Uniquely compresses analytics and archival data with hybrid columnar compression.
- More Users: Uniquely prioritizes CPU and I/O by user, job, service, session, and SQL.
Oracle has also done some interesting engineering work to allow customers to control how much power each Exadata database server consumes. Oracle enables customers to cap power usage and dynamically reduce power consumption during periods of low usage. These benefits will be top of mind for customers deploying systems in their own data centers since they must pay for all ancillary costs and source their own power, so cutting power consumption both saves money and helps them meet sustainability goals.
Moreover, with Exadata administration efficiency I find that organizations can spend less by managing less. Exadata automates platform hardening, tuning, and securing alongside Exadata Cloud automating infrastructure management and Autonomous Database automating database management. These are essential capabilities to enable customers to take full advantage of the decisive Exadata X11M benefits in key areas such as performance, scaling, memory, energy efficiency gains, and multi-cloud agility.
Looking Ahead: Exadata X11M Ushers in Next-Generation Intelligent Data Architecture
I believe it has never been clearer that Exadata is the industry standard platform to run Oracle Database. Whether it’s powering Oracle Database 23ai in hyperscalers such as AWS, Azure, and Google Cloud—or outperforming on-premises infrastructure from any major vendor out there—the latest Exadata X11M allows organizations to decide where they want to gain the best performance for their Oracle Database workloads.
And for everything from extreme vector processing to stock exchange-level transactions, Exadata X11M is the gold standard by which all other platforms that try to run Oracle Database are measured. Trying to run Oracle Database on anything other than Exadata and expecting a superior result is analogous to trying to wear sandals while trying to outrun Erriyon Knighton – it’s just not happening. Save the Dell, HPE, Pure Storage, and generic compute cloud instances for the secondary app tier, and leave the heavy lifting of Oracle Database workloads to Exadata X11M.
See the official Oracle press release on the Oracle site.
Disclosure: The Futurum Group is a research and advisory firm that engages or has engaged in research, analysis, and advisory services with many technology companies, including those mentioned in this article. The author does not hold any equity positions with any company mentioned in this article.
Analysis and opinions expressed herein are specific to the analyst individually and data and other information that might have been provided for validation, not those of The Futurum Group as a whole.
Other insights from The Futurum Group:
OCW24: Oracle Unveils OCI ZPR to Simplify Cloud Security
Oracle and Google Cloud Finally Tie the Interconnect Knot
Oracle Exadata Exascale: Making Data Architecture for Cloud Intelligent
Author Information
Ron is an experienced, customer-focused research expert and analyst, with over 20 years of experience in the digital and IT transformation markets, working with businesses to drive consistent revenue and sales growth.
He is a recognized authority at tracking the evolution of and identifying the key disruptive trends within the service enablement ecosystem, including a wide range of topics across software and services, infrastructure, 5G communications, Internet of Things (IoT), Artificial Intelligence (AI), analytics, security, cloud computing, revenue management, and regulatory issues.
Prior to his work with The Futurum Group, Ron worked with GlobalData Technology creating syndicated and custom research across a wide variety of technical fields. His work with Current Analysis focused on the broadband and service provider infrastructure markets.
Ron holds a Master of Arts in Public Policy from University of Nevada — Las Vegas and a Bachelor of Arts in political science/government from William and Mary.