#intelxeoncpus

Improve Intel Data Direct I/O (DDIO) Workload Performance with Intel VTune Profiler.

Profile uncore hardware performance events in Intel Xeon processors with oneAPI

One hardware feature included in Intel Xeon CPUs is Intel Data Direct I/O (DDIO) technology. By making the CPU cache the primary point of entry and exit for I/O data going into and out of the Intel Ethernet controllers and adapters, it contributes to advances in I/O performance.

To monitor the effectiveness of DDIO and Intel Virtualization Technology (Intel VT) for Directed I/O (Intel VT-d), which permits the independent execution of several operating systems and applications, it is essential to monitor uncore events, or events that take place outside the CPU core. By analyzing uncore hardware events, you may improve the performance of Intel Data Direct I/O (DDIO) workloads using Intel VTune Profiler, a performance analysis and debugging tool driven by the oneAPI.

We’ll talk about using VTune Profiler to evaluate and enhance directed I/O performance in this blog. Let’s take a quick look at Intel Data Direct I/O technology before we go into the profiling approach.

Overview of the Intel Data Direct I/O (DDIO) Technology

Intel Integrated I/O technology Intel DDIO was launched in 2012 for the Intel Xeon processor E5 and E7 v2 generations. It aims to increase system-level I/O performance by employing a new processor-to-I/O data flow.

I/O operations were sluggish and processor cache was a scarce resource prior to the development of Data Direct I/O technology. It was necessary for the host processor’s main memory to store and retrieve any incoming or departing data from an Ethernet controller or adapter, respectively. It used to be necessary to move the data from main memory to the cache before working with it.

This led to a lot of read and write operations in the memory. This also caused some additional, speculative read operations from the I/O hub in some of the older designs. Excessive memory accesses often lead to higher system power consumption and deterioration of I/O performance.

Intel DDIO technology was created to rearrange the flow of I/O data by making the processor cache the primary source and destination of I/O data instead of the main memory, as the processor cache is no longer a restricted resource.

Depending on the kind of workload at the workstation or on the server, the DDIO approach offers benefits like:

Higher transaction rates, reduced battery usage, reduced latency, increased bandwidth, and more.

There is no industry enablement needed for the Data Direct I/O technology.

It doesn’t rely on any hardware, and it doesn’t need any modifications to your operating system, drivers, or software.

Boost DDIO Performance Using Intel VTune Profiler

A function carried out in a CPU’s uncore section, outside of the processor core itself, that yet affects processor performance as a whole is referred to as an uncore event. For instance, these occurrences may be connected to the Intel Ultra Path Interconnect (UPI) block, memory controller, or I/O stack action.

A new recipe in the VTune Profiler Cookbook explains how to count these kinds of uncore hardware events using the tool’s input and output analysis function. You may analyze Data Direct I/O and VT-d efficiency by using the data to better understand the traffic and behavior of the Peripheral Component Interconnect Express (PCIe).

The recipe explains how to do input and output analysis, evaluate the findings, and classify the resulting I/O metrics. In essence, VTune Profiler v2023.2 or later and an Intel Xeon scalable CPU of the first or later generation are needed. Although the approach is suitable to the most recent version of Intel Xeon Processors, the I/O metrics and events covered in the recipe are based on the third generation Intel Xeon Scalable Processor.

Perform I/O Analysis with VTune Profiler

Start by analyzing your application’s input and output using VTune Profiler. With the analysis function, you may examine CPU, bus, and I/O subsystem use using a variety of platform-level metrics. You may get data indicating the Intel Data Direct I/O(DDIO) use efficiency by turning on the PCIe traffic analysis option.

Analyze the I/O Metrics

VTune Profiler Web Server or VTune Profiler GUI may be used to examine the report that is produced as a consequence of the input and output analysis. Using the VTune Profiler Web Server Interface, the recipe illustrates the examination of many I/O performance indicators, including:

Platform diagram use of the physical core, DRAM, PCIe, and Intel UPI linkages.

PCIe Traffic Summary, which includes metrics for both outgoing (caused by the CPU) and incoming (caused by I/O devices) PCIe traffic.

These measurements aid in the computation of CPU/IO conflicts, latency for incoming read/write requests, PCIe bandwidth and efficient use, and other factors.

Metrics to assess the workload’s effectiveness in re-mapping incoming I/O device memory locations to various host addresses using Intel VT-d technology.

Usage of DRAM and UPI bandwidth.

The range of IBM  Cloud Bare Metal Servers for Virtual Private Clouds is being shaken up by new 4th Gen Intel Xeon processors and dynamic network bandwidth.

With great pleasure, IBM is thrilled to announce that the fourth generation of Intel Xeon CPUs are now available on IBM  Cloud Bare Metal Servers for Virtual Private Clouds. IBM customers now have the ability to provision Intel’s most recent micro architecture within their very own virtual private cloud. This allows them to get access to a variety of performance benefits, such as increased core-to-memory ratios (21 new server profiles) and dynamic network bandwidth that is only available through IBM Cloud VPC. For those who are following track, that is three times as many provisioning options as their present Intel Xeon  CPUs, which are of the second generation. Take a look around.

Are these servers made of bare metal suitable for my needs?

In addition to having rapid provisioning, excellent network speeds, and the most secure software-defined resources that are accessible within IBM, IBM  Cloud Bare Metal Servers for Virtual Private Clouds are hosted on their most recent and developer-friendly platform. Every single one of your central processing units would be based on the 4th gen Intel Xeon processors, which IBM initially introduced on IBM  Cloud Bare Metal Servers for traditional infrastructure in conjunction with Intel’s day-one release product.

The traditional IBM Cloud infrastructure is distinct from the IBM Cloud Virtual Private Cloud. More suitable for large, steady-state, predictable activities that call for the highest possible level of customisation is this method. However, IBM Cloud Virtual Private Cloud is an excellent solution for high-availability and maximum elasticity requirements. Take a look at this brief introduction video to get a better understanding of which environment would be most suitable for your workload requirements.

The customisation choices available to you include five pre-set profile families, which contain your number of  CPU instances, RAM, and bandwidth, in the event that IBM  Cloud Bare Metal Servers for Virtual Private  Cloud turns out to be your preferred choice. What sets IBM Cloud apart from other cloud services is the fact that each profile provides you with DDR-5 memory and dynamic network bandwidth ranging from 10 to 200 Gbps. For tasks that require a significant amount of  CPU power, such as heavy web traffic operations, production batch processing, and front-end web servers, compute profiles are the most effective solution.

Balanced profiles are designed to provide a combination of performance and scalability, making them a great choice for databases of a moderate size and cloud applications that experience moderate traffic.

Memory profiles are most effective when applied to workloads that require a significant amount of memory, such as large cache and database applications, as well as in-memory analytics.

When it comes to running small to medium in-memory databases and OLAP, such as SAP BW/4 HANA, very high profiles are the most effective solutions.

Large in-memory databases and online transaction processing workloads are both excellent for ultra-high profiles because they offer the most memory per core.

For these bare metal servers, what kinds of workloads do you propose they handle?

Over the course of this year, IBM’s beta programme was exposed to a wide variety of workloads; nonetheless, there were a few noteworthy success stories that particularly stood out:

Building on top of IBM  Cloud, VMware Cloud Foundation These workloads required a high core performance, interoperability with VMware, licencing portability, a smaller core count variety, and a Generic operating system, which IBM just recently launched. In a dedicated location, they conducted tests for VMware managed virtual cloud functions (VCF) as well as build-your-own VMware virtual cloud functions (VCF).

They were happy with the customisation freedom and benchmark performance enhancements that backed up their findings. During the second half of the year, these workloads will be accessible on Intel Xeon profiles of the fourth generation within the IBM  Cloud Virtual Private Cloud.

With regard to HPCaaS, this workload was one of a kind, and IBM believe that it is a primary use case for this distribution. Terraform and IBM Storage Scale were used in their tests to see whether or not they could get improved performance. They were delighted with the throughput improvement and the agile provisioning experience between platforms and networking.

The task of providing financial services and banking necessitated both powerful and dedicated system performance, as well as the highest possible level of security and compliance. After conducting tests to determine capacity expansion, user interface experience, security controls, and security management, they were thrilled to find that production times had been reduced.

Beginning the process

In the data centres of IBM  Cloud Dallas, Texas, bare metal servers powered by 4th gen Intel Xeon processors are currently accessible. Additional sites will be added in the second half of the year 24. The IBM  Cloud Bare Metal Servers for Virtual Private Cloud catalogue allows you to view all of the pricing and provisioning options for their new 4th Gen Intel Xeon processors and save a quote to your account. As an alternative, you could start a chat and obtain some answers right now. Within their cloud documents, you can find more information by reading their getting started guides and tutorials.

Spend one thousand dollars in IBM Cloud credits

If you are an existing customer who is interested in provisioning new workloads or if you are inquisitive about deploying your first workload on IBM Cloud VPC, then you should be sure to take advantage of their limited time promotion for IBM Cloud VPC. By entering the promotional code VPC1000 within either the bare metal or virtual server catalogues, you will receive USD 1,000 in credits that may be used towards the purchase of your new virtual private cloud (VPC) resources. These resources include computing, network, and storage components. Only profiles based on the second generation of Intel Xeon processors and profiles from earlier generations are eligible for this promotion, which is only available for a limited period.