Not All Clouds Are Created Equal

Why the performance you demand can also deliver the most value.

Succeeding with the cloud requires prioritizing – and investing for – long-term planning over short-term ‘quick wins’. The good news for IT leaders, however, is that high-performing technology can also deliver the most value. You need your cloud on Intel.1

Creating the right IT environment to support a competitive business requires careful planning and preparation. In this context, it can be tempting to look only at short-term considerations, such as upfront costs. However, thinking through the longer-term total cost of ownership (TCO) of your Amazon Web Services (AWS) cloud instances ensures you are considering the broadest possible range of factors - including performance and overall value.

Strategic hardware upgrades boost performance – and save money

Cutting-edge technologies offer new, higher levels of performance. For example, AWS instances based on the latest generation of Intel® Xeon® Scalable processors offer the highest levels of performance compared to instances based on previous generation technology, with 2x faster reliable performance for analytics and database processing.2

At a first glance, choosing a public cloud platform based on the latest technology may seem costly, but research has shown that performance improvements and size reduction mean the opposite. In fact, organizations can reduce per-instance infrastructure costs by moving to smaller, newer instance sizes.3 For example, an enterprise using AWS EC2 T2.Large instances can upgrade to AWS EC2 T3.Medium instances to achieve a saving of USD 248.14 per instance without losing any throughput or performance. This will result in an annual saving of USD 32,258.3

In addition to this efficiency saving, reducing the number of cores in the data center often brings down software licensing costs, which could represent another large saving. It’s estimated that enterprises who upgrade to the newest AWS EC2 instances could save as much as USD 72,000 annually.4

Prioritize value over cost by comparing performance-per-dollar

The value of an AWS EC2 instance is more than just its upfront cost. When you’re assessing different options, look into benchmarking and performance-per-dollar for the workloads that matter most to your enterprise and business outcomes. Demanding workloads like high performance computing (HPC) and artificial intelligence (AI) are transforming many business use cases, from unlocking new levels of medical research to advanced fraud detection in the financial industry.

Consider the performance-per-dollar of different AWS instances specifically for running the workloads that matter to your enterprise. You may find that while less expensive instances look like a better deal, in many cases Intel® technology provides higher value where it counts - anywhere from 1.25x greater performance for web/Java workloads all the way up to 4.1x greater performance per dollar with high performance computing (HPC).5

Build a seamless IT environment to move data with ease

Your AWS cloud services are just one part of your IT infrastructure. To ensure you can flexibly move data and workloads around your digital estate with minimal latency, you’ll need to ensure architecture compatibility. Because Intel® architecture-based cloud instances are completely compatible, you don’t need to shut down the workload and restart it on the new architecture. This means faster, smoother, and more efficient transfer of workloads and data with no downtime.

When it comes to picking the AWS cloud solution that is best suited to your needs, there are many different factors to consider. Speed and performance are important regardless of your workload – and by measuring TCO over upfront cost, you can ensure these factors are fully integrated into your company’s cloud strategy.

Download Intel’s business brief to find out more about building a TCO-focused business strategy.

Product and Performance Information


Software and workloads used in performance tests may have been optimized for performance only on Intel® microprocessors. Performance tests, such as SYSmark* and MobileMark*, are measured using specific computer systems, components, software, operations, and functions. Any change to any of those factors may cause the results to vary. You should consult other information and performance tests to assist you in fully evaluating your contemplated purchases, including the performance of that product when combined with other products. For more information go to

Performance results are based on testing as of the date set forth in the configuration details and may not reflect all publicly available security updates. See configuration disclosure for details. No product or component can be absolutely secure.


Up to 4.2x more VMs based on server virtualization consolidation workload: Based on Intel internal estimates 1-Node, 2x Intel® Xeon® processor E5-2690 on Romley-EP with 256 GB Total Memory on VMware ESXi* 6.0 GA using Guest OS RHEL 6.4, glassfsh3.1.2.2, postgresql9.2. Data Source: Request Number: 1718, Benchmark: Server utilization consolidation, Score: 377.6 @ 21 VMs vs 1-Noble, 2 x Intel® Xeon® Platinum 8180 processor on Wolf Pass SKX with 768 GB of Total Memory on VMware ESXi6.0 U3 GA using Guest OS RHELL 6 64bit. Data Source: Request Number: 2563. Benchmark: Server virtualization consolidation. Score: 1580 @ 90 VMs. Higher is better. URL TBC – TSOlogic research report, New advances by Intel and Amazon Web Services, Drive Major Cloud Savings.


Savings figures and percentages are derived from analysis conducted on TSO Logic’s 100,000-instance repository of anonymized data. To conduct this analysis, the TSO Logic platform algorithmically analyzed a subset of our 100,000-instance repository of on-premise OS instances across companies evaluating cloud. The platform created a fine-grained statistical model of each organization’s compute resources to determine the most cost-effective place to run each workload. Ingesting millions of data points from the current environments—including age, generation, and configuration of all hardware, the OSs they’re running, and each instance’s utilization—it algorithmically profiled compute patterns. It then used machine learning influenced algorithms and pattern matching to determine the best fit for each workload from thousands of potential cloud options. Using up-to-date, validated information from Intel and AWS, the platform normalized and compared processing capabilities and costs across AWS instance types using various generations of Intel® Xeon® processors. Cost reduction scenarios described are intended as examples of how a given Intel®-based product, in the specified circumstances and configurations, may affect future costs and provide cost savings. Circumstances will vary. Intel and AWS does not guarantee any costs or cost reduction.URL TBC – TSOlogic research report, New advances by Intel and Amazon Web Services, Drive Major Cloud Savings.


Instance change: C4.8XLarge to C5.4XLarge, cost savings per instance $3,270.33, percentage saved per instance 50%. Moving from C4.8XLarge instances to C5.4XLarge reduces the core count for that workload by 40. For each such C4 instance the customer migrates to a C5, they will realize up to $3,270 infrastructure savings. But, if the customer is running Microsoft* SQL Server on those instances, which is licensed at a cost of $1,800 per core annually, the core count reduction lowers their licensing costs by $72,000. URL TBC – TSOlogic research report, New advances by Intel and Amazon Web Services, Drive Major Cloud Savings.


Results as of January 4th, 2019. Performance testing done on AWS EC2 M and R instances ( Database Benchmarks show R5 vs. R5a due to higher memory capacity.

Memory capacity for instances: M5.24xlarge and m5a.24xlarge: 384GB; r5.24xlarge and r5a.24xlarge: 768GB.

For details on EC2 instance protections for various vulnerabilities including side-channel, please refer to Amazon security bulletins: