Most organizations understand the value of moving certain applications to the cloud as part of a smart multi-cloud strategy – determining not only which applications would benefit most from being cloud-based but also defining what kinds of instances to run them on. A vast range of instance options are offered, but their underlying technology and the performance gains they deliver are not always clearly understood. Too often, decision makers will focus primarily on per hour or per second compute cost when in fact a different instance might return a more favorable price performance.
One misperception is that taking the time to analyze and select newer instance options is simply too burdensome for busy IT and business professionals with schedules to keep and competing demands to satisfy. As a result, often the choice is made to stick with the status quo – if things are working just fine now, why change? Or, it might be that your cloud strategist simply is unaware of the latest available instance options and their potential impact on your workload performance, resulting in missed opportunities to improve performance and reduce costs.
Presidio believes now is the time to reassess the instances you’re using in your cloud strategy with the availability of instances using the latest generation of Intel® Xeon® Scalable processors, which are boosting performance in the cloud at a price point equal to, or even lower than, prior generation instances.
Extend your cloud instance advantages with newer Intel instance types
Cloud instances based on 3rd Gen Intel® Xeon® Scalable processors have been optimized for a diverse range of modern workload types from leading software providers and offer a balanced architecture with built-in acceleration and advanced security capabilities. For example:
- Crypto Acceleration enhances data protection and privacy by increasing the performance of encryption-intensive workloads, such as Secure Sockets Layer (SSL) web serving, 5G infrastructure, and Virtual Private Networks (VPN) firewalls.
- Intel® Deep Learning Boost (Intel® DL Boost) speeds up complex artificial intelligence (AI) inferencing and training workloads and can provide more cost-effective inferencing performance than relying upon expensive GPUs.
- Intel® Advanced Vector Extensions 512 (Intel® AVX-512) boosts performance and throughput for the most demanding computational tasks in applications such as modeling and simulation, data analytics and machine learning, data compression, visualization, and digital content creation.
- Intel® Software Guard Extensions (Intel® SGX) protects data and application code within multi-tenant public cloud environments via application isolation in memory, independent of operating system or hardware configuration.
These underlying technologies can have a significant impact on cost to performance as cloud instances run today’s demanding applications faster and more efficiently, contributing to a greater return on investment.
Understanding the value of built-in crypto acceleration
Cryptographic operations are among the most critical but compute-intensive processes required to move, manipulate and store data securely. Research from TechRepublic revealed that over the past few years many organizations have adopted encryption in the public cloud – yet the dangers from increased cyber-threats and the demands of exploding data traffic have continued to drive the need for ever-greater compute cycles for crypto processing.
Intel addressed that challenge with the April 2021 launch of the 3rd Gen Intel Xeon Scalable processor featuring Intel crypto acceleration. Optimized for cloud, enterprise, high-performance computing and other demanding workloads, its advanced algorithm offers up to 4.2X greater cryptographic performance wherever sensitive data is at rest or on the move – in databases, 5G infrastructure, XaaS environments, SSL web serving and more – all while mitigating the performance impact of pervasive encryption.
Previously this kind of crypto acceleration was available only as an add-on capability at extra cost. Now, crypto acceleration is built into select cloud instances with the 3rd Gen Intel Xeon Scalable processor, with Intel® Total Memory Encryption (Intel® TME) providing full physical memory encryption as part of a complete security feature set.
In short, the advantages of built-in crypto acceleration make these new instances a “must-have” for any environment that uses crypto functions to protect critical data anywhere from edge to data center to multi-tenant public cloud while it is stored or processed.
New cloud instance workloads that benefit from the latest Intel technologies
Consider just a few real-world examples of environments that are already achieving significant performance gains from the 3rd Gen Intel Xeon Scalable processors:
- Azure Virtual Desktop (AVD): As remote and hybrid work models grow in acceptance and organizations race to accommodate this ascendent mobile labor force, the need for virtual desktop infrastructure (VDI) will continue to expand – and so will the technology to support it. AVD based on Dv5 Azure VMs is able to accommodate up to 12 percent more VDI users with the 3rd Gen Intel Xeon scalable processors, an ideal choice for general-purpose workloads delivering up to 15 percent greater performance at an improved price-to-performance ratio.
- General compute: The above-mentioned usage model focuses on the growing need for secure, high-performance virtualization. But in truth, just about any workload from edge to cloud can derive measurable, bottom-line benefits by reassessing one’s cloud strategy to take advantage of the pervasive cryptographic capabilities and other performance enhancements built into 3rd Gen Intel Xeon Scalable processors for greater performance at an equal or lower price, compared to previous generations.
- Machine learning on Amazon EC2 DL1 instances: Beyond the processor, compute-intensive workloads for artificial intelligence (AI), machine learning and deep learning require the right acceleration technologies to accomplish their mission while also protecting their critical data. Amazon EC2 DL1 instances, aided by Gaudi accelerators from Intel-owned Habana Labs, offer deep learning models training for natural language processing, object detection and image recognition, all at a low cost. DL1 instances offer as much as 40 percent performance improvement for training deep learning models, versus the metrics for previous-gen GPU-based EC2 instances – and up to 60 percent better price-performance over EC2 P4d instances (8 x A100 GPUs).
The key to getting the most from these critical technologies is to start with a serious reassessment of your cloud workload instances and what they’re running on, as well as how you might benefit from the new instances and their built-in features. And the best part is, you don’t have to go it alone.
We can help you determine your idea instance fit – today and tomorrow
To ensure that you’re deriving the most value from your cloud strategy, it’s important to understand your current instances and the potential benefits you could achieve from newer instance types. Fortunately, your assessment process need not be complicated or time-consuming.
Presidio is pleased to help perform an audit of your cloud instances for actionable information, insight and recommendations: detailing your current cloud strategy, identifying instances and software choices that could derive the greatest price-performance value, and sharing what’s coming. Our goal is to position you to take advantage of future technology innovations, simplifying the next iteration of your cloud migration path.
Don’t settle for today’s status quo at the risk of missing out on a performance boost from 3rd Gen Intel Xeon Scalable processors with built-in cryptographic acceleration on key instances, augmented by other Intel technologies. Contact Presidio to start rethinking your cloud strategy.
Copyright © 2022 IDG Communications, Inc.