Top benefits to using Intel Optane NVMe for cache drives in VxRail
Wed, 20 May 2020 14:42:17 -0000|
Read Time: 0 minutes
Performance, endurance, and all without a price jump!
There is a saying that “A picture paints a thousand words” but let me add that a “graph can make for an awesome picture”.
Last August we at VxRail worked with ESG on a technical validation paper that included, among other things, the recent addition of Intel Optane NVMe drives for the vSAN caching layer. Figure 3 in this paper is a graph showing the results of a throughput benchmark workload (more on benchmarks later). When I do customer briefings and the question of vSAN caching performance comes up, this is my go-to whiteboard sketch because on its own it paints a very clear picture about the benefit of using Optane drives – and also because it is easy to draw.
In the public and private cloud, predictability of performance is important, doubly so for any form of latency. This is where caching comes into play, rather than having to wait on a busy system, we just leave it in the write cache inbox and get an acknowledgment. The inverse is also true. Like many parents I read almost the same bedtime stories to my young kids every night, you can be sure those books remain close to hand on my bedside “read cache” table. This write and read caching greatly helps in providing performance and consistent latency. With vSAN all-flash there no longer any read cache as the flash drives at the capacity layer provide enough random read access performance… just as my collection of bedtime story books has been replaced with a Kindle full of eBooks. Back to the write cache inbox where we’ve been dropping things off – at some point, this write cache needs to be empty, and this is where the Intel Optane NVMe drives shine. Drawing the comparison back to my kids, I no longer drive to a library to drop off books. With a flick of my finger I can return, or in cache terms de-stage, books from my Kindle back to the town library - the capacity drives if you will. This is a lot less disruptive to my day-to-day life, I don’t need to schedule it, I don’t need to stop what I’m doing, and with a bit of practice I’ve been able to do this mid story Let’s look at this in actual IT terms and business benefits.
To really show off how well the Optane drives shine, we want to stress the write cache as much as possible. This is where benchmarking tools and the right knowledge of how to apply them come into play. We had ESG design and run these benchmarking workloads for us. Now let’s be clear, this test is not reflective of a real-world workload but was designed purely to stress the write cache, in particular the de-staging from cache to capacity. The workload that created my go-to whiteboard sketch was the 100% sequential 64KB workload with a 1.2TB working set per node for 75 minutes.
The graph clearly shows the benefit of the Optane drives, they keep on chugging at 2,500MB/sec of throughput the entire time without dropping a beat. What’s not to like about that! This is usually when the techie customer in the room will try to burst my bubble by pointing out the unrealistic workload that is in no way reflective of their environment, or most environments… which is true. A more real-world workload would be a simulated relational database workload with a 22KB block size, mixing random 8K and sequential 128K I/O, with 60% reads and 40% writes, and a 600GB per node working set, which is quite a mouthful and is shown in figure 5. The results there show a steady 8.4-8.8% increase in IOPS across the board and a slower rise in latency resulting in a 10.5% lower response time under 80% load.
Those of you running OLTP workloads will appreciate the graph shown in figure 6 where HammerDB was used to emulate the database activity of a typical online brokerage firm. The Optane cache drives under that workload sustained a remarkable 61% more transactions per minute (TPM) and new orders per minute (NOPM). That can result in significant business improvement for an online brokerage firm who adopts Optane drives versus one who is using NAND SSDs.
When it comes to write cache, performance is not everything, write endurance is also extremely important. The vSAN spec requires that cache drives be SSD Endurance Class C (3,650 TBW) or above, and Intel Optane beats this hands down with an over tenfold margin at 41 PBW (41,984 TBW). The Intel Optane 3D XPoint architecture allows memory cells to be individually addressed in a dense, transistor-less, stackable design. This extremely high write endurance capability has let us spec a smaller sized cache drive, which in turn lets us maintain a similar VxRail node price point, enabling you the customer to get more performance for your dollar.
What’s not to like? Typically, you get to pick any two; faster/better/cheaper. With Intel Optane drives in your VxRail you get all three; more performance and better endurance, at roughly the same cost. Wins all around!
Author: David Glynn, Sr Principal Engineer, VxRail Tech Marketing
Related Blog Posts
VxRail brings key features with the release of 4.7.510
Thu, 18 Jun 2020 14:24:47 -0000|
Read Time: 0 minutes
VxRail recently released a new version of our software, 4.7.510, which brings key feature functionality and product offerings
At a high level, this release further solidifies VxRail’s synchronous release commitment with vSphere of 30 days or less. VxRail and the 4.7.510 release integrates and aligns with VMware by including the vSphere 6.7U3 patch release. More importantly, vSphere 6.7U3 provides the underlying support for Intel Optane persistent memory (or PMem), also offered in this release.
Intel Optane persistent memory is non-volatile storage medium with RAM-like performance characteristics. Intel Optane PMem in a hyperconverged VxRail environment accelerates IT transformation with faster analytics (think in-memory DBMS), and cloud services.
Intel Optane PMem (in App Direct mode) provides added memory options for the E560/F/N and P570/F and is supported on version 4.7.410. Additionally, PMem will be supported on the P580N beginning with version 4.7.510 on July 14.
This technology is ideal for many use cases including in-memory databases and block storage devices, and it’s flexible and scalable allowing you to start small with a single PMem module (card) and scale as needed. Other use cases include real-time analytics and transaction processing, journaling, massive parallel query functions, checkpoint acceleration, recovery time reduction, paging reduction and overall application performance improvements.
New functionality enables customers to schedule and run "on demand” health checks in advance, and in lieu of the LCM upgrade. Not only does this give customers the flexibility to pro-actively troubleshoot issues, but it ensures that clusters are in a ready state for the next upgrade or patch. This is extremely valuable for customers that have stringent upgrade schedules, as they can rest assured that clusters will seamlessly upgrade within a specified window. Of course, running health checks on a regular basis provides sanity in knowing that your clusters are always ready for unscheduled patches and security updates.
Finally, the VxRail 4.7.510 release introduces optimized security functionality with two-factor authentication (or 2FA) with SecurID for VxRail. 2FA allows users to login to VxRail via the vCenter plugin when the vCenter is configured for RSA 2FA. Prior to this version, the user would be required to enter username and password. The RSA authentication manager automatically verifies multiple prerequisites and system components to identify and authenticate users. This new functionality saves time by alleviating the username/password entry process for VxRail access. Two-factor authentication methods are often required by government agencies or large enterprises. VxRail has already incorporated enhanced security offerings including security hardening, VxRail ACLs and RBAC, KMIP compliant key management, secure logging, and DARE, and now with the release of 4.7.510, the inclusion of 2FA further distinguishes VxRail as a market leader.
Please check out these resources for more VxRail 4.7.510 information:
By: KJ Bedard - VxRail Technical Marketing Engineer
Big Solutions on Dell EMC VxRail with SQL 2019 Big Data Cluster
Thu, 09 Jul 2020 19:20:04 -0000|
Read Time: 0 minutes
The amount of data and different formats organizations must manage, ingest, and analyze has been the driving force behind Microsoft SQL 2019 Big Data Clusters (BDC). SQL Server 2019 BDC demonstrates the deployment of scalable clusters of SQL Server, Spark, and containerized HDFS (Hadoop Distributed File System) running on Kubernetes.
We recently deployed and tested SQL Server 2019 BDC on Dell EMC VxRail hyperconverged infrastructure to demonstrate how VxRail delivers the performance, scalability, and flexibility needed to bring these multiple workloads together.
The Dell EMC VxRail platform was selected for its ability to incorporate compute, storage, virtualization, and management in one platform offering. The key feature of the VxRail HCI is the integration of vSphere, vSAN, and VxRail HCI System Software for an efficient and reliable deployment and operations experience. The use of VxRail with SQL Server 2019 BDC makes it easy to unite relational data with big data.
The testing demonstrates the advantages of using VxRail with SQL Server 2019 BDC for analytic application development. This also demonstrates how Docker, Kubernetes, and the vSphere Container Storage Interface (CSI) driver accelerate the application development life cycle when they are used with VxRail. The lab environment for development and testing used four VxRail E560F nodes supported by the vSphere CSI driver. With this solution, developers can provision SQL Server BDC in containerized environments without the complexities of traditional methods for installing databases and provisioning storage.
Our white paper, Microsoft SQL Server 2019 Big Data Cluster on Dell EMC VxRail shows the power of implementing SQL Server 2019 BDC technologies on VxRail. Integrating SQL Server 2019 RDBMS, SQL Server BDC, MongoDB, and Oracle RDBMS helps to create a unified data analytics application. Using VxRail enhances the ability of SQL Server 2019 to scale out storage and compute clusters while embracing the virtualization techniques from VMware. This SQL Server 2019 BDC solution also benefits from the simplicity of a complete yet flexible validated Dell EMC VxRail with Kubernetes management and storage integration.
The solution demonstrates the combined value of the following technologies:
- VxRail E560F – All-flash performance
- Large tables stored on a scaled-out HDFS storage cluster that is hosted by BDC
- Smaller related data tables that are hosted on SQL Server, MongoDB, and Oracle databases
- Distributed queries that are enabled by the PolyBase capability in SQL Server 2019 to process Transact-SQL queries that access external data in SQL Server, Oracle, Teradata, and MongoDB.
- Red Hat Enterprise Linux
Big Data Cluster Services
This diagram shows how the pools are built. It provides details of the benefits for Kubernetes features for container orchestration at scale, including:
- Autoscaling, replication, and recovery of containers
- Intracontainer communication, such as IP sharing
- A single entity—a pod—for creating and managing multiple containers
- A container resource usage and performance analysis agent, cAdvisor
- Network pluggable architecture
- Load balancing
- Health check service
This white paper, Microsoft SQL Server 2019 Big Data Cluster on Dell EMC VxRail, addresses big data storage, the tools for handling big data, and the details around testing with TPC-H. When we tested data virtualization with PolyBase, the queries were successful, running without error and returning the results that joined all four data sources.
Because data virtualization does not involve physically copying and moving the data (so that the data is available to business users in real-time), BDC simplifies and centralizes access to and analysis of the organization’s data sphere. It enables IT to manage the solution by consolidating big data and data virtualization on one platform with a proven set of tools.
Success starts with the right foundation:
SQL Server 2019 BDC is a compelling new way to utilize SQL Server to bring high-value relational data and high-volume big data together on a unified, scalable data platform. All of this can be deployed with VxRail, enabling enterprises to experience the power of PolyBase to virtualize their data stores, create data lakes, and create scalable data marts in a unified, secure environment without needing to implement slow and costly Extract, Transform, and Load (ETL) pipelines. This makes data-driven applications and analysis more responsive and productive. SQL Server 2019 BDC and Dell EMC VxRail provide a complete unified data platform to deliver intelligent applications that can help make any organization more successful.
Read the full paper to learn more about how Dell EMC VxRail with SQL 2019 Big Data Clusters can:
- Bring high-value relational data and high-volume big data together on a single, scalable platform.
- Incorporates intelligent features and gets insights from more of your data—including data stored beyond SQL Server in Hadoop, Oracle, Teradata, and MongoDB.
- Supports and enhances your database management and data-driven apps with advanced analytics using Hadoop and Spark.
Additional VxRail & SQL resources:
Author: Vic Dery, Senior Principal Engineer, VxRail Technical Marketing