Datacenter Real World Workloads are very different from synthetic lab workloads and can be used to compare performance of various types of storage. SQL Server workloads for a Retail Web Portal and a Navigation Portal are examined and used as test stimulus to compare datacenter storage at the block IO level. A comparison is provided for Persistent Memory and various classes of solid state storage.

N/A

A lot has changed since DiskSpd was first open sourced in 2014. This session aims to provide you with an insight into several of the newer features of the tool and discuss ways in which you can contribute to the project. The session will also drill down on the use of the Windows Performance Analyzer to understand DiskSpd generated I/Os and performance in a Windows environment.

What if we had a library that could encapsulate ATA, SCSI & NVMe command set, encapsulate the OS level nuances of Windows, Linux, FreeBSD & Solaris and can compile not only on x86 but on ARM, MIPS & PowerPC etc. as well? Well there is one and best thing of all, it’s open source.

There has been a tremendous rise in the computing power of handheld devices over the last 5 years or so. What used to be extremely limited device that was best used for making and receiving calls and text messages (anybody remember the pager), are today more powerful than computers that helped in taking Apollo 11 to the moon.

Byte-addressable non-volatile memory (PMEM) offers persistence at near DRAM memory speeds in large enough capacity to change the storage vs memory paradigm used by databases, analytics, and filesystems. We believe virtualization is fundamental to workload agility and consolidation for these new applications. We will provide an overview of VMware virtualization for PMEM that is now running on real PMEM products.

The enterprise applications use huge amount of data which demands large scale distributed storage subsystems deployments in the data centers. The storage virtualization functionalities bring in more complexity in storage management. The latest storage systems support performance data collection with high frequencies such as seconds and minutes which enables to gather more data for performance analysis. The challenge with the storage subsystems is finding the performance bottlenecks and identifying the root cause and resolving them with less turnaround.

When running a file server, a lot of bytes transferred back and forth aren’t necessarily needed by the server itself. User data read from files and sent over the network doesn’t need to be copied to the user space, just so the server process can send it, which is the idea behind the “zero-copy transfer”. The IO vector is a concept extending it by comprising both user and kernel space data in various possible arrangements.

This presentation explores some design decisions around enhancing the zfs send and zfs receive commands to transfer already compressed data more efficiently and to recover from failures without re-sending data that has already been received.

Learning Objectives

High level understanding - how ZFS provides an efficient platform for async replication
Finding stability in the chaos - tension between what's stable in an archive and what isn't
Resolving significant constraints - why something simple turned out not to be not so simple

For years, enterprises have desired a way to ensure that an application will operate properly at scale on a particular storage array and network infrastructure, and increasingly, ensure that multiple applications can safely run at capacity on an all-flash or hybrid architecture. Companies need an approach that enables users to reference detailed workload analysis profiles as they build and share workload models, and simulate performance across a variety of vendors, products and technologies.

Learning Objectives