EMC Symmetrix

Marketing Hype:

The Symmetrix system is EMC Corporation's flagship enterprise storage array. The Symmetrix development has been led by Moshe Yanai, who joined EMC in 1987, until shortly before his leaving EMC in 2001. There have been many generations of Symmetrix hardware, with the first appearing in 1990 and the latest, the Symmetrix VMAX Series, introduced in April 2009. The development of the Symmetrix's new features and huge internal software base, amid increasing its capacity and performance in orders of magnitude, have been an on-going process since its introduction.

History

Symmetrix arrays, EMC's flagship product at that time, began shipping in 1990 as an IBM Mainframe Block Multiplex Channel - (aka Bus and Tag) connected storage array. Newer generations of Symmetrix brought additional host connection protocols which include ESCON, SCSI, Fibre Channel-based SANs, FICON and iSCSI. The Symmetrix product was initially popular within the airline industry and with companies that were willing to deviate from the safety of IBM's 3390 disk subsystem and take a risk with the unproven Symmetrix array. The Symmetrix has been developed by a team led by Moshe Yanai. This product is the main reason for the rapid growth of EMC in the 1990s, both in size and value, from a company valued hundreds of millions of dollars to a multi-billion company. Moshe Yanai managed the Symmetrix development from the product's inception in the late 1980s until shortly before leaving EMC in 2001, and his Symmetrix development team grew from several people to thousands.

In the mid-1990s the Symmetrix expanded beyond mainframes (Bus and Tag, ESCON) into open systems (SCSI), and replication features such as SRDF (remote replication) and TimeFinder (local replication) have been added to it. Other storage interface types (e.g., Fibre channel, FICON, iSCSI) have been added upon their introduction in the market. Since the appearance of Storage Area Networks (SANs), Fibre Channel first, and then Ethernet based, the Symmetrix has been a leader in this area.

Symmetrix VMAX Series

EMC Symmetrix VMAX systems are enterprise-class storage platforms intended for open systems and mainframe computing. VMAX systems run the Enginuity operating environment for Symmetrix and are offered in two models:

• The scalable Symmetrix VMAX system
• The single engine Symmetrix VMAX SE system.

The Symmetrix VMAX System

The Symmetrix VMAX system is the high-end, scalable storage array with a system bay and separate roll-up storage bays. The system scales from a single Symmetrix VMAX Engine system with one storage bay to a large eight engine system and a maximum of ten storage bays. System upgrades are achieved by adding single or multiple VMAX Engines or additional storage bays. Each VMAX Engine contains two Symmetrix VMAX directors with extensive CPU processing power, physical memory, front-end ports, and back-end ports. Drive capacity is increased by installing 4 Gb/s disk array enclosures to the storage bay.

The Symmetrix VMAX SE System

The Symmetrix VMAX SE system is the single engine Symmetrix system. Symmetrix VMAX SE systems offer a single cabinet configuration that contains both the system logic and drives.

Symmetrix DMX Series

The DMX family of Symmetrix disk arrays was first introduced in February 2003. This new line of arrays replaced the old direct connect line of Symmetrix with an array whose components were all interconnected together. This release delivered the first modular Symmetrix which was the DMX800 and well as the monolithic DMX1000. Later revisions of the family were named DMX2, DMX3 and DMX4.

Platform Comparisons

	Symmetrix VMAX	Symmetrix VMAX SE	Symmetrix DMX-4	Symmetrix DMX-4 950
Maximum Drives	2400	360	2400	360
Architecture	Virtual Matrix Architecture	Virtual Matrix Architecture	Direct Matrix Architecture	Direct Matrix Architecture
Maximum Integrated Directors	16	2	N/A	N/A
Connection Types	FC, FICON, Gigabit Ethernet, iSCSI	FC, FICON, Gigabit Ethernet, iSCSI	FC, FICON, ESCON, Gigabit Ethernet, iSCSI	FC, Gigabit Ethernet, iSCSI
Maximum Connectivity	Up to 128 ports depending on connection type	Up to 16 ports depending on connection type	Up to 64 ports depending on connection type	Up to 16 ports depending on connection type
Maximum SRDF Ports	32 (FC or GbE SRDF ports)	4 (FC or GbE SRDF ports)	8 (GbE SRDF ports)	8 (GbE SRDF ports)

Enginuity Software

Enginuity is the name of the storage operating environment that controls components in an EMC Symmetrix storage array. This specialized operating environment is optimized for data storage functions. It is driven by real-time events related to the input and output of data and applies self-optimizing intelligence to deliver performance, availability and data integrity for EMC advanced storage arrays.

Advancements in Enginuity are carried forward in each successive EMC platform generation. This means that all of the reliability, availability, and serviceability features; all of the interoperability and host operating systems coverage; and all of the application software capabilities developed by EMC and its partners continue to perform productively and seamlessly as underlying hardware technology is refreshed.

Enginuity encompasses four key concepts:

• Foundation—Preemptive multi-tasking, operational consistency, and security. Managing multiple shared resources across Symmetrix systems, Enginuity provides built-in security capabilities while insulating storage applications from technology changes.
• Performance—Maximizing speed. Utilizing patented intelligent adaptive algorithms to manage data flow across channels, memory, and disks, Enginuity dynamically controls events in complex and highly variable environments to maximize application performance under varying workloads.
• Availability—Always accessible data. Enginuity manages data integrity through redundant functions in hardware and software including mirrored memory, redundant disk placement and infrastructure to assure data availability and reliability. Availability and reliability includes trend analysis and early detection as well as automatic failover and fully automated escalation upon problem detection.
• Open integration—Comprehensive coverage, guaranteed interoperability, and investment protection.

EMC maintains a storage networking interoperability program for hardware and software. In addition, using openly available application programming interfaces (APIs) and supporting SMI industry standards, independent software vendors are able to utilize Enginuity functions in their own packages.

Enginuity on Symmetrix provides multiple RAID protection levels as well as copies of data locally, on the same array, or remotely on another remote Symmetrix array with EMC’s Symmetrix Remote Data Facility (SRDF). Importantly, Enginuity can control the consistency of data copies at various locations allowing for multiple application uses as well as full security and data integrity in the event of a catastrophic shutdown at any client site. Enginuity also includes capabilities like Fully Automated Storage Tiering (FAST) and Virtual Provisioning – EMC’s thin provisioning option.

Symmetrix Software

TimeFinder

TimeFinder is a family of EMC replication products that operate in a single Symmetrix array and non-disruptively create and manage point-in-time copies of data volumes. TimeFinder runs in Symmetrix Enginuity but is controlled by TimeFinder software running on an attached host. It can be administered by the user through Solutions Enabler Command Line Interface, Symmetrix Management Console (SMC), EMC Control Center (ECC) or Mainframe Enablers. TimeFinder includes the following sets of products:

TimeFinder/Mirror

TimeFinder/Mirror is the original TimeFinder product that has been in existence for about 12 years. It provides full copies of source volumes through a technique of hardware mirroring. The target volume for a TimeFinder/Mirror process is a Business Continuance Volume (BCV); a specially designated volume within the Symmetrix configuration. When a BCV is fully synchronized with a data device, the BCV is separated or split, and made available to a host for backup or other host processes.

TimeFinder/Clone

TimeFinder/Clone provides single or multiple point-in-time copies of full volumes or individual datasets. Cloned data is available to a host immediately upon activation, even if the copy process has not completed.

TimeFinder/Snap

TimeFinder/Snap provides pointer-only based replicas simultaneously on multiple target devices from a single source device. With a space saving TimeFinder/Snap only changed data is written to a pool of save devices. Data reconstruction is from the source device and the pointers into the change tracking save pool. Data may be copied from a single source device to as many as 128 target virtual devices. Because it uses pointers, the additional capacity to support the copy is minimal –typically less than 30% of the source volume.

TimeFinderClone and Snap are fully compatible with each other. For example, when using TF/Clone with space saving TF/Snap, our customers can solve critical business issues in many different ways, supporting a wide range of service-level objectives.These copies can be used to facilitate non-disruptive backups, decision support, application, testing, and development, third party software updates, and even large data migration projects.

TimeFinder Consistency Groups

TimeFinder Consistency Groups is a flow control mechanism that is employed to create copies of source volumes at an instantaneous point in time. TF/CG utilizes an Enginuity feature called ECAM (Enginuity Consistency Assist) to momentarily halt writes to source volumes during the creation of the copy.

SRDF

The Symmetrix SRDF (Symmetrix Remote Data Facility) family of remote mirroring software offers various levels of Symmetrix based business continuity and disaster recovery solutions. The SRDF product family offers the capability to maintain multiple, host-independent, mirrored copies of data. The Symmetrix systems participating in SRDF can be in the same room, in different buildings within the same campus, or hundreds to thousands of kilometers apart.

SRDF transparently remotely mirrors production or primary (source) site data to a secondary (target) site to users, applications, databases, and host processors. The local SRDF device, known as the primary (RDF1) device, is configured in a partner relationship with a remote secondary (RDF2) device to form an SRDF pair. By maintaining copies of data in different physical locations, SRDF enables the following operations with minimal impact on normal business processing:

• Disaster restart
• Disaster restart testing
• Recovery from planned outages
• Remote backup
• Data center migration
• Data replication and mobility

Upon the previously described functionalities, SRDF may be implemented in several modes:

• a synchronous version called SRDF/S
• a consistent asynchronous version called SRDF/A
• a triangular solution called SRDF/Star

Advanced IP features

IPv6

Historic network standard IPv4 is still supported with the introduction of the advanced multiprotocol front-end channel directors. IPv6 differs from IPv4 in that it offers a much larger IP addressing scheme in the form of 128-bit allocated blocks. The available 32-bit IPv4 addressing space of 4 billion addresses became sparse quickly during the Internet surge and became a concern. Another way of addressing for the Internet standards needed to be created. The largest concern of having limited addressing is the potential ease of IP duplication throughout the global environment and the additional threat of easier IP spoofing across networks. Because IPv4 had limitations that resulted in reactive solutions to extensibility issues with security, multicasting, CIDR, NAT, quality service, and mobility, the standardization of IPv6 had more baselines to create proactive solutions to existing addressing problems. Using CIDR (classless interdomain routing) and NAT (network address translation) allowed IPv4 administrators to carve up environments into smaller manageable segments and translating private addressing to global address translation became more scalable. The advancement of IPv6 has eliminated the need for these management tools providing unique node capabilities and driven more peer-to-peer applications like SRDF and iSCSI to utilize peer-to-peer “globally routable” addressing without the worry about administration of NAT or subnets. The benefits of using and managing IPv6 goes beyond its larger, diverse address space capabilities. Even though EMC software cannot take advantage of the “autoconfiguration” feature in IPv6, the ease of plugging an IPv6 device into a network environment and obtaining a global address easily is a huge advantage to mobility. This isn’t to say that other features don’t complement the EMC software capabilities. For example, neighbor discovery with a routable environment provides the available IPv6 links to “discover” one another without the need to know the subnet mask or static route of the environment.

SRDF and iSCSI connection requirements will be the same over a customer’s IPv6 network with functionality equivalent to IPv4. The SRDF sessions must run within an IPv6 network and transfer data over IPv6 networks. They require entry of IPv6 addresses in all GUIs where an IPv4 address can be entered. SRDF and iSCSI will use IPv6 versions of any IPv4 services as they are currently available. These include DNS, DHCP, and others. The only exceptions are the variables that make IPv6 distinct from IPv4, which are that there are no subnet masks or static routes needed.

IPv6 configuration requirements are as follows:

• GigE configurations require the EMC customer engineer to configure static parameters through the Symmwin GUI onsite.
• SRDF specific configurations require the EMC customer engineer to input IP Address specifics through the Symmwin GUI onsite.
• Customers need to provide IP address schemas between sites for implementation by an EMC customer engineer.
• Autoconfiguration of IPv6 addressing is not allowed. Some customers may use this feature that allows devices (like a router) to generate their own globally routable addresses as needed. This conflicts with the requirement that addressing remains static on the Symmetrix (neighbor discovery).
• Switches and routers in the path of IPv6 traffic must be IPv6 capable.

IPSec

The advanced multiprotocol front-end channel directors for Symmetrix provide the same level of security as standalone encryption appliances while minimizing management and reliability overhead associated with external appliances. They handle IPsec by using the integrated Hifn 4450 processor chips. Having IPSec capabilities in the Symmetrix is desirable for customer wanting to secure their SRDF transactions across IP networks via Gigabit Ethernet.

IPsec is the security protocol designed to encapsulate the entirety of IP data over a network between hardware endpoints. It has the capacity to combine strong authentication and complementary encryption algorithms to create secure associations (SAs) between local and remote entities. IPsec operates at Layer 3 (network layer) of the OSI reference model and has been integrated into IPv6 extension headers. This has reduced the dependency on hardware vendors from having to “hook” this protocol into the IP stack for secure connectivity. This has not, however, changed for the IPv4 standard. The entirety of IPsec is made up of the Authentication Header (AH), Encapsulating Security Protocol (ESP), and Internet Key Exchange (IKE). The minimum criterion for IPsec is to use ESP and IKE with preshared keys. This implementation is primarily what the advanced multiprotocol boards will use with Enginuity 5773.

IPsec uses IKE inside an Internet Security Association Key Management Protocol (ISAKMP) framework to negotiate protocols and algorithms based on the policies set up locally on the entity. ISAKMP is the local policy generated to dictate the encryption and authentication methodology in how the preshared keys will be used by the endpoint negotiation for a secure session. These policies will be set up through EMC Solutions Enabler (CLI) or Symmetrix Management Console. However, if customers want advanced administration features such as external key manager integration or hardened key storage, they will need to rely on appliances for site-to-site connectivity. External key management is being currently evaluated to provide a complete public-key infrastructure (PKI) by RSA with EMC to provide “extra strength” encryption using public/private key exchange. The difference between PKI and preshared keys used for IPsec is that a complete public key infrastructure uses a trusted certificate authority that creates a public/private digital key pair to create an IPsec connection. Even though this method is more secure, it is more difficult to manage than preshared keys. In the meanwhile, preshared keys are the default method of session key establishment for Enginuity 5773. External encryption devices have been qualified historically for connectivity between sites; however, this does not provide hardware encryption end to end, as shown in Figure

IPsec typically does not impact performance or latency. There is the initialization phase during which the first security association is established. Subsequent sessions will be seamless as connections remain during the policy allocated lifetime. Unlike some other encryption implementations, the encryption is not performed in software, but rather in an embedded line-grade encryption co-processor, and it does not degrade performance. With encryption technologies such as SSL, heavy TCP applications behave sluggish and overhead can impact round-trip turnaround time between the source and destination.

Quality of Service

Two Quality of Service tools available to EMC Symmetrix storage arrays are Dynamic Cache Partitioning (DCP) and Symmetrix Priority Controls (SPC). Storage resource optimization based on workload equality is the default behavior for Symmetrix arrays catering to homogenous application environments. However consolidation of dissimilar workloads on the same storage resources has become more desirable and this goal requires flexibility for differential treatment between workflows. Cache memory and disk access are key storage resources now subject to isolation and prioritization mechanisms. The ability to set allocation preferences for these resources facilitates many storage management objectives.

Dynamic Cache Partitioning

With Dynamic Cache Partitioning (DCP) workloads that are cache friendly can be guaranteed more cache resources, increasing the aggregate cache utilization figure and improving that workload’s performance. A working set that is not cache effective can be fenced into a small cache partition, removing any cache diluting influence for other applications. The resulting increase in aggregate cache utilization improves the effective use of cache and allows maximal performance for prioritized workloads.

DCP allows for cache partitions of fixed size, partitions that shrink and grow based on cache efficiency measures and the scheduling of cache partition size changes to meet the needs of changing business cycles.

Symmetrix Priority Controls

Many strategies currently exist to deliver improved performance for disk drives within storage subsystems. Disk level buffering and request reordering are techniques that have evolved to specifically address disk optimization opportunities inherent in storage throughput operations. Faster seek times, higher disk spin rates and reduced interface overheads attempt to counterbalance disk capacity increases and deliver improved access performance. All of these techniques function well in homogenous environments. However, Symmetrix Priority Controls (SPC) is designed to provide distinction between multiple application workloads by setting preferences for higher-tier applications, during times of disk contention. SPC is designed to function only when disk contention is discovered. Implementing user requested priorities at the disk level will deliver:

• Higher I/O rates on a disk while maintaining response time for priority work
• Consistent response times for priority workloads
• Balanced disk utilization through workload peaks and troughs
• Greater disk utilization
• More effective use of larger capacity, lower performance disks
• Simultaneous low priority work with minimal contention on higher priority work

Virtual Provisioning

Virtual Provisioning is EMC’s implementation of thin provisioning. Virtual Provisioning allows users to create large “thin” volumes and present them to the host while consuming physical storage from a shared pool only as needed.

Symmetrix Virtual Provisioning introduces a new type of host accessible device called a thin device that can be used in many of the same ways that regular, host accessible Symmetrix devices have traditionally been used. Unlike regular Symmetrix devices, thin devices do not need to have physical storage completely allocated at the time the devices are created and presented to a host. The physical storage that is used to supply drive space for a thin device comes from a shared thin storage pool that has been associated with the thin device.

A thin storage pool is composed of a new type of internal Symmetrix device called a data device that is dedicated to the purpose of providing the actual physical storage used by thin devices.

When a write is performed to a portion of the thin device, the Symmetrix allocates a minimum allotment of physical storage from the pool and maps that storage to a region of the thin device including the area targeted by the write. The storage allocation operations are performed in small units of storage called “thin device extents.” A round-robin mechanism is used to balance the allocation of data device extents across all of the data devices in the pool that are enabled and that have remaining unused capacity. The thin device extent size is twelve 64 KB tracks (768 KB).

When a read is performed on a thin device, the data being read is retrieved from the appropriate data device in the storage pool to which the thin device is bound.

Virtual Provisioning thin devices are supported for use with all Open Systems platforms that are qualified for connectivity to EMC Symmetrix DMX and VMAX disk arrays. EMC Symmetrix Fully Automated Storage Tiering (FAST), automates tiered storage strategies by moving workloads between Symmetrix tiers as performance characteristics change over time. FAST performs system reconfiguration, improving performance and reducing costs, while maintaining service levels.

Fully Automated Storage Tiering (FAST)

Designed to work in thick and thin provisioned environments, FAST automates the identification of data volumes for the purposes of relocating application data across different performance/capacity tiers within an array. FAST proactively monitors workloads at the volume level in order to identify “busy” volumes that would benefit from being moved to higher performing drives. FAST also identifies less “busy” volumes that could be relocated to higher capacity drives, without existing performance being affected. This promotion/demotion activity is based on policies that associate a storage group to multiple drive technologies, or RAID protection schemes, based upon the performance requirements of the application contained within the storage group. Data movement executed during this activity is performed non-disruptively, without affecting business continuity and data availability.

FAST uses three distinct algorithms when determining the appropriate tier for a device. The algorithms, in order of probability, are:

• EFD promotion/demotion algorithm
• Capacity-based algorithm
• FC/SATA cross-tier algorithm

The goal of the EFD promotion/demotion algorithm is to maximize Flash drive utilization within the array. When complete, the algorithm will have listed all the devices in the array in order of which devices would be best served being configured on EFD. FAST will then attempt to place those devices onto Flash drives. The goal of the capacity-based algorithm is to enforce the FAST policy storage usage rules. A storage group is considered to be in violation when a higher percentage of devices exist on a tier than is configured in the policy for that tier. The goal of the FC/SATA cross-tier algorithm is to balance utilization across Fibre Channel and SATA technologies. Devices are sorted by disk service time, and the most utilized devices will be moved to the least utilized disks. If Optimizer is also enabled on the Symmetrix, then the traditional Optimizer algorithm will be used to balance load within a physical disk group.

There are two methods by which a device will be relocated to another tier: swap or move. A swap occurs when there is no unconfigured space in the target tier, and results in a corresponding device being moved out of the target tier. In order to preserve data on both devices involved in the swap, a single DRV is used. A move occurs when unconfigured space exists in the target tier. Only one device is involved in a move, and a DRV is not required. Symmetrix metadevices are moved as a complete entity—metadevice members may not exist in different physical disk groups.

Management and operation of FAST is provided by EMC Symmetrix Management Console (SMC), as well as the EMC Solutions Enabler Command Line Interface (SYMCLI).

Virtual LUN

Symmetrix Virtual LUN Technology enables tiered storage strategies by allowing manual “re-tiering” of data as its value changes over time. Symmetrix Virtual LUN assists with system reconfiguration, performance improvement and consolidation efforts while maintaining service levels. Virtual LUN technology, enhanced with Enginuity 5874 for the Symmetrix VMAX Series, enables transparent, nondisruptive data mobility among storage tiers within the same array and between RAID protection schemes.

Virtual LUN technology offers two types of data movement: migration to unconfigured space and migration to configured space. In each case, the migration provides users the ability to move data between high-performance drives and high-capacity drives, or to populate newly added drives, with full inter-RAID flexibility.

Virtual LUN technology is supported for both open system and mainframe devices, and includes support for metavolumes. Virtual LUN technology is fully interoperable with all other Symmetrix replication technologies – SRDF, TimeFinder/Clone, TimeFinder/Snap, and Open Replicator. Virtual LUN migrations can be managed via the Symmetrix Management Console (SMC) graphical interface, or the Solutions Enabler Command Line Interface (SYMCLI).

Management Software

EMC provides host packages that to assist in the management of the Symmetrix array. These host packages perform functions today that used to be the exclusive domain of EMC Customer Engineers many years ago. The following sections list the host software that provide management controls over the Symmetrix:

Solutions Enabler

EMC Solutions Enabler is software that provides a host with SYMAPI (Symmetrix Application Programming Interface), CLARAPI (CLARiiON Application Programming Interface) and STORAPI (Storage Application Programming Interface) shared libraries for use by Solutions Enabler applications. Solutions Enabler was developed by EMC for storage, system, and database administrators, and systems engineers. It provides a specialized library of UNIX-formatted commands, and supports command line entries and scripts to perform configuration, control, and management operations on devices and data objects in EMC storage environments. Solutions Enabler software supports both open systems and mainframe operating systems.

Example Solutions Enabler operations;

• Set array-wide metrics.
• Control operations on array devices and ports.
• Device creation, device provisioning (for host allocation), and creation of device pools for thin provisioning.
• Fully automated storage tiering (FAST).
• Optimize array performance.
• Manage QOS (Quality of Service) metrics.
• Perform virtual LUN migration tasks.

Solutions Enabler software is accompanied with every EMC storage array (Symmetrix/CLARiiON) that is sold. This is a critical piece of software and fully compatible with either Symmetrix Enginuity or CLARiiON Flare. Enginuity and Flare are the intelligent underlying software that drives the storage array operating platform.

Symmetrix Management Console

Symmetrix Management Console (SMC) is a browser-based user interface designed for configuring and managing Symmetrix arrays. It was developed to concurrently support all the features of Enginuity Version 5671 and later.

SMC presents the functionality of the Solutions Enabler SYMCLI (command line interface) in a browser interface. SMC is used to perform the following functions;

• Manage Symmetrix access controls, user accounts, and permission roles
• Discover Symmetrix arrays
• Perform configuration operations (create devices, map and mask devices, set Symmetrix attributes, set device attributes, set port flags, create SAVE device pools)
• Manage devices (change device configuration, set device status, reserve devices, duplicate devices, create/dissolve metadevices)
• Manage Fully Automate Storage Tiering (FAST), virtual provisioning, and Auto-provisioning Groups
• Perform and monitor replication operations (TimeFinder/Mirror, TimeFinder/Snap, TimeFinder/Clone, SRDF, Open Replicator)
• Monitor alerts
• Monitor an application’s performance

Symmetrix Management Console can manage storage related operations from device creation and provisioning to features such as, FAST, replication configuration and monitoring. SMC manages physical and virtual storage. SMC operates on Symmetrix arrays in Mainframe, Open Systems and iSeries environments. SMC manages up to ten Symmetrix DMX and/or VMAX units, and up to 80,000 storage volumes.

SMC deploys a client/server model where it can be installed in local or remote locations. In a local installation, the SMC software is installed on the same system as the SYMAPI (Symmetrix Application Programming Interface) server. In a remote installation, the SMC software is installed on a system that is connected to the SYMAPI server. The ‘SYMAPI server’ and ‘base’ license keys are required as part of the installation for use with SMC, and a Java runtime environment must be enabled on the client browser. SMC is also available as a virtual appliance for the ESX v3.5 (and later) in the VMware infrastructure.

Symmetrix Performance Analyzer

EMC Symmetrix Performance Analyzer (SPA) is a browser-based tool used to perform historical trending and analysis of Symmetrix array performance data. SPA was developed to work with the Symmetrix Management Console (SMC). The SPA interface can open in its own web window from the SMC menu, or on its own. SPA adds an optional layer of data collection, analysis, and presentation tools to the SMC implementation. You can use SPA to:

• Set performance thresholds and alerts
• View high frequency metrics as they become available
• Perform root cause analysis
• View graphs detailing system performance
• Drill down through data to investigate issues
• Monitor performance and capacity over time

SPA also provides a “fast lane” to display possible performance road blocks with one click, and includes export and print capability for all data graphs.

Symmetrix Features for Mainframe

Symmetrix provides specific features that provide compatibility with mainframe storage arrays provided by IBM. These features are listed below:

Parallel Access Volume (PAV)

Parallel Access Volumes are implemented within z/OS allowing one I/O to take place for each base unit control block (UCB), and one for each statically or dynamically assigned alias UCB. These alias UCBs allow parallel I/O access for volumes. Current Enginuity releases provide support for static, dynamic, and hyperPAVs. HyperPAVs allow fewer aliases to be defined within a logical control unit. With hyperPAVs, aliases are applied to the base UCBs (devices) when the need arises.

Multiple Allegiance (MA)

While PAVs facilitate multiple parallel accesses to the same device from a single LPAR, Multiple Allegiance (MA) allows multiple parallel nonconflicting accesses to the same device from multiple LPARs. Multiple Allegiance I/O executes concurrently with PAV I/O. The Symmetrix storage system treats them equally and guarantees data integrity by serializing writes where extent conflicts exist.

FlashCopy Support

Symmetrix arrays running Enginuity 5772 and earlier support FlashCopy channel commands through use of z/OS host emulation package that runs as a started task. Symmetrix arrays running Enginuity 5773 and later have support in Enginuity itself. In other words, a FlashCopy channel command is sent to the array where the command is executed by Enginuity.

ESCON support

Enterprise Systems Connection (ESCON) is a fiber-optic connection technology that interconnects mainframe computers, workstations and network-attached storage devices across a single channel, and supports half duplex data transfers. ESCON may also be used for handling Symmetrix Remote Data Facility (SRDF) remote links.

FICON support

Fiber Connection (FICON) is a fiber-optic channel technology that extends the capabilities of its previous fiber optic channel standard, ESCON. Unlike ESCON, FICON supports full duplex data transfers and enables greater throughput rates over longer distances. FICON uses a mapping layer based on technology developed for Fibre Channel and multiplexing technology, which allows small data transfers to be transmitted at the same time as larger ones. With Enginuity release level 5670 and later, Symmetrix storage systems support FICON ports. With the Enginuity service release 5874.207, VMAX supports 8 Gb FICON connectivity (FICON Express8).

Fibre Channel support

Fibre Channel is a supported option with z/VM and z/Linux.

Extended Address Volume

The ability to utilize volumes that are greater than 65,000 cylinders was provided in z/OS 1.10. EMC Symmetrix VMAX arrays utilizing Enginuity 5874.207 support EAVs.

zHPF support

System z10 High Performance FICON (zHPF) represents the latest enhancement to the FICON interface architecture aimed at offering an improvement in the performance of online transaction processing (OLTP) workloads. Mainframe systems that are presently channel-constrained running heavy workloads using a 4K page size will reap the greatest benefit from this feature.

External links

• SAN vs DAS: A Cost Analysis of Storage in the Enterprise
• EMC.com
• EMC Symmetrix V-Max and the Virtual Data Center
• Learn EMC Storage Product
• Utilities to create EMC Symmetrix/Vmax Configuration Scripts