December 16, 2013 at 13:49Dell Powervault MD3 Storage Arrays
Sooner or later, many companies are faced with the need to store an ever-increasing amount of work information. Thus, according to studies, in 2011 the total amount of data created reached 1.8 trillion GB, and according to Gartner forecasts, the total data growth by 2016 will amount to 850%. Moreover, almost everywhere, in addition to storing such data, the ability to quickly access and / or change them is required. The average data growth for the year may exceed 100% and therefore the requirements for storage grow faster than for the computing power of servers. In addition, the increase in disk capacity sometimes leads to unsatisfactory results from the use of traditional RAID schemes.
Dell’s portfolio offers a one-stop solution for this: PowerVault MD3 high-density storage.
The PowerVault MD3 product line can be conditionally divided into groups according to the form factor of the disks used (2.5 ”or 3.5”), their type of connection to the existing infrastructure (SAS, iSCSI, FC) and expansion options. Due to the scalability of the solution, the total number of disks used by the system can reach 180 units in the high-density disk enclosure (4U) version, or 192 in configurations with regular-density disk enclosures (2U).
All key components of the system (controllers, power supplies, disks, fans) are duplicated and can be replaced in the hot-swap mode, and the controllers are backed up by high-capacity capacitors to be able to write the contents of the cache memory to non-volatile memory in case of external power failure. Among the undoubted advantages of this solution is the ability to keep the cache data consistent for any long time, a much longer life of the capacitors compared to batteries and the absence of the need for preventive replacement on average every three years. High-density arrays (4U or MD3x60 models) have separate independent shelves with which you can replace or add disks without stopping application I / O.
In addition to standard RAID levels 0, 1, 10, 5, and 6 for standard storage systems, the MD3 family of arrays offers the user a separate option for managing disks — dynamic pools or Dynamic Disk Pools (DDP). The DDP pool dynamically distributes data, free capacity, and protection (parity) information in the disk pool. Pools can be a minimum of 11 disks to all available disks in the PowerVault MD3 storage system, including expansion disk enclosures. In addition to creating one or more dynamic pools, storage administrators can create traditional disk RAID groups in conjunction with dynamic pools, providing the highest level of flexibility.
DDP pools greatly simplify storage administration, as there is no need to manage idle resources or many separate RAID groups.
Dynamic Disk Pools are composed of several low-level elements. The first one is called D-Piece. It consists of a 512 MB continuous partition on a physical disk containing 4096 128 KB segments. From the selected disks in the pool, 10 D-Piece components are selected using an intelligent optimization algorithm. 10 connected D-Piece elements form a 4 GB D-Stripe. The contents of the D-Stripe component are similar to the RAID 6 partition in the 8 + 2 configuration, where 8 base segments can contain user data, 1 segment contains parity (P) information derived from user data segments, and the last segment contains the Q value according to RAID 6.
Virtual logical disks or LUNs (Logical Unit Number) are created, in essence, by combining several 4-GB D-Stripe elements to obtain the specified size of the virtual disk within the maximum permissible size in the Dynamic Disk Pool.
When the storage administrator has finished defining the Dynamic Disk Pool, which for the most part consists of setting the required number of disks in the pool, D-Piece and D-Stripe structures are created, similar to traditional RAID blocks when creating a virtual disk. After defining Dynamic Disk Pool in the pool, you can create a virtual disk. This disk will consist of several D-Stripe elements located on all disks in the pool, with a total capacity of up to a specified value of the virtual disk capacity, for example, the number of D-Stripe = selected capacity / 4 GB. You can define many virtual disks in the Dynamic Disk Pool, and you can create several pools in the storage system itself.
The storage administrator can also create traditional RAID RAID groups together with Dynamic Disk Pools or any combination of them. Like traditional RAID disk groups, Dynamic Disk Pools can be expanded by adding disks to the pool during Dynamic Capacity Expansion (DCE). You can add up to 12 disks to a pool at a time. When the DCE operation is initiated, part of the existing D-Piece elements that form the logical volumes is actually transferred to the new disks, thereby increasing the number of physical disks that host the logical volume and improving performance.
Another important advantage of the Dynamic Disk Pool is that instead of using dedicated isolated components to hot-swap failed disks, the pool itself has a built-in backup capacity on each disk included in the pool in case of possible failures. This simplifies management, as you no longer need to plan or monitor individual dedicated hot-swap drives. At the same time, recovery time is significantly reduced (two to three times when using large disks), and the performance of logical volumes during recovery is practically not reduced compared to traditional dedicated hot-swap disks.
In addition, given that virtual disks in Dynamic Disk Pools can contain a very large number of disks compared to traditional RAID 5 or RAID 6 disk groups with up to 30 disks per group, then in environments with mixed or non-parallel loads, you can achieve benefits in terms of performance, since all pool resources are available to all logical volumes at once.
In addition to the use of DDP, due to the possibility of combining SSDs and traditional hard drives in one solution, it became possible to implement caching on SSDs. SSD caching allows the PowerVault MD3 array to use solid-state drives as an extended read cache, thereby improving application performance with a large number of arbitrary read operations, for example for file servers, web servers, databases, etc. The
solid state drive cache is optional and is used along with the core to improve performance. When creating a cache on an SSD, it is divided into two internal virtual RAID 0 disks (one disk per controller). This cache volume is not available for normal data storage.
In full duplex mode, each controller uses a separate virtual disk, gaining access to only half of the SSD space, even if the controller fails or enters maintenance mode. In simplex mode, one controller manages the entire SSD space and uses both virtual disks. The maximum allowable cache size on a solid-state drive in PowerVault MD3 arrays is 4 TB.
SSD caching provides benefits for workloads with the following features:
SSD cash is recommended when a balanced, cost-effective approach is required using a combination of hard drives and cache on a solid state drive, and the cost of dedicated SSD drives for individual logical volumes is prohibitive.
Another feature of PowerVault MD3 storage is synchronous and asynchronous remote replication. It allows you to effectively replicate data, protecting information from system crashes and power failures. This feature duplicates data between PowerVault MD3 storage systems to ensure availability in the event of a failure, as well as for other business data migration purposes.
In the synchronous model, one array stores the primary virtual disk, and the second, remote, stores the secondary virtual disk of the replication pair. After establishing a connection between the primary and secondary arrays, the primary array processes the server-initiated write operations on the secondary virtual disk, sending write requests to the secondary array, while simultaneously performing these operations on the local array. The primary array returns to the server system a successful result only after both arrays have completed the write operation.
The synchronous model has several advantages, the most important of which is the ability to recover application data from a secondary array, including the latest updates. If a critical failure occurs on the primary array, then application data can be restored from the secondary array, then only write requests that were executed during the failure will disappear. Since the application on the side of the primary array expects a positive response about the completion of the current requests, their loss will not cause discrepancies in the data in the event of a server failure or reboot during their execution. Applications that have even the most basic recovery capabilities, in such cases, are able to restore data integrity on their own. Consequently, Using a synchronous remote replication model allows applications to realize fault tolerance. The disadvantages of synchronous replication include high requirements for communication channels.
Asynchronous remote replication is a great choice for those who need protection against failures, but who do not have the means to implement synchronous replication models. It is assumed that such consumers do not need an absolutely reliable recovery plan until the failure, they are more interested in the cost-effectiveness of the solution. In other words, asynchronous replication allows the consumer to use the slower and less expensive communication channels between the primary and secondary arrays if certain delays are allowed, which are not present when using the synchronous replication model. Asynchronous replication model allows you to adjust the duration of the delay depending on the needs of the client in accordance with the capabilities and characteristics of the communication channel. One of the main benefits of asynchronous remote replication is that that the performance of the application in the main array will not be related to the communication channel. Of course, channel performance will affect latency, but it will not directly affect application performance.
Asynchronous remote replication is designed to work with relatively slow and low-speed communication channels (for example, TCP / IP channels for iSCSI). The main mode of its operation includes customizable periods of divergence, during which no attempts are made to transfer data from the primary array to the secondary one. Instead, using a dirty card allows the array to track the changed portions of the primary virtual disk for subsequent selective transfer to the secondary array. After the time period specified in the settings has expired, the primary array starts transferring the corresponding data to the secondary array. This approach provides several advantages, each of which is especially important when using slow, global communication channels.
Asynchronous Remote Replication for Dell Storage Systems PowerVault MD3 provides a valuable solution for customers who need to provide fault tolerance to protect critical business data without the significant financial investment that requires synchronous remote replication. Asynchronous remote replication can be used on relatively slow communication channels, allowing the user to provide a reasonable compromise between RPO (Return Point Objective) and the cost of communication lines. Therefore, asynchronous remote replication allows you to transmit data over long distances via TCP / IP networks, providing protection against disasters, but without significantly affecting the performance of applications on the primary node.
In addition, using efficient MD Storage Manager software provides simple, easy-to-use, task-based configuration and monitoring tools. Add to this the VMware vStorage APIs for array integration (VAAI, VASA support), support for operating systems WS2008, Windows 2012, RHEL 5.8, 6.2, SLES 10.4, 11.2, Citrix XenServer, VMware ESXi 5.0, 5.1, MS Hyper-V 2008, work with thin volumes and you get a reliable system for using the most modern and most resource-intensive applications. Based on the requirements for the system, you can use the 90-day trial license to use the following features:
Activation is done through a graphical interface (MD Storage Manager) and automatically turns off after 90 days of use. At the same time, working data is not lost, but becomes inaccessible until the license is purchased.
Thus, PowerVault MD3 storage systems can find their application in a wide range of tasks. From general-class tasks (consolidation of business applications, virtualization projects, archive storage systems, etc.), to high-performance computing applications (computing clusters, video surveillance, digital content distribution).
For additional information on the configuration, capabilities or ordering conditions for the PowerVault MD3 family of storage systems, you can contact Marat Rakayev , the Dell Moscow office.
Dell’s portfolio offers a one-stop solution for this: PowerVault MD3 high-density storage.
The PowerVault MD3 product line can be conditionally divided into groups according to the form factor of the disks used (2.5 ”or 3.5”), their type of connection to the existing infrastructure (SAS, iSCSI, FC) and expansion options. Due to the scalability of the solution, the total number of disks used by the system can reach 180 units in the high-density disk enclosure (4U) version, or 192 in configurations with regular-density disk enclosures (2U).
All key components of the system (controllers, power supplies, disks, fans) are duplicated and can be replaced in the hot-swap mode, and the controllers are backed up by high-capacity capacitors to be able to write the contents of the cache memory to non-volatile memory in case of external power failure. Among the undoubted advantages of this solution is the ability to keep the cache data consistent for any long time, a much longer life of the capacitors compared to batteries and the absence of the need for preventive replacement on average every three years. High-density arrays (4U or MD3x60 models) have separate independent shelves with which you can replace or add disks without stopping application I / O.
In addition to standard RAID levels 0, 1, 10, 5, and 6 for standard storage systems, the MD3 family of arrays offers the user a separate option for managing disks — dynamic pools or Dynamic Disk Pools (DDP). The DDP pool dynamically distributes data, free capacity, and protection (parity) information in the disk pool. Pools can be a minimum of 11 disks to all available disks in the PowerVault MD3 storage system, including expansion disk enclosures. In addition to creating one or more dynamic pools, storage administrators can create traditional disk RAID groups in conjunction with dynamic pools, providing the highest level of flexibility.
DDP pools greatly simplify storage administration, as there is no need to manage idle resources or many separate RAID groups.
Dynamic Disk Pools are composed of several low-level elements. The first one is called D-Piece. It consists of a 512 MB continuous partition on a physical disk containing 4096 128 KB segments. From the selected disks in the pool, 10 D-Piece components are selected using an intelligent optimization algorithm. 10 connected D-Piece elements form a 4 GB D-Stripe. The contents of the D-Stripe component are similar to the RAID 6 partition in the 8 + 2 configuration, where 8 base segments can contain user data, 1 segment contains parity (P) information derived from user data segments, and the last segment contains the Q value according to RAID 6.
Virtual logical disks or LUNs (Logical Unit Number) are created, in essence, by combining several 4-GB D-Stripe elements to obtain the specified size of the virtual disk within the maximum permissible size in the Dynamic Disk Pool.
When the storage administrator has finished defining the Dynamic Disk Pool, which for the most part consists of setting the required number of disks in the pool, D-Piece and D-Stripe structures are created, similar to traditional RAID blocks when creating a virtual disk. After defining Dynamic Disk Pool in the pool, you can create a virtual disk. This disk will consist of several D-Stripe elements located on all disks in the pool, with a total capacity of up to a specified value of the virtual disk capacity, for example, the number of D-Stripe = selected capacity / 4 GB. You can define many virtual disks in the Dynamic Disk Pool, and you can create several pools in the storage system itself.
The storage administrator can also create traditional RAID RAID groups together with Dynamic Disk Pools or any combination of them. Like traditional RAID disk groups, Dynamic Disk Pools can be expanded by adding disks to the pool during Dynamic Capacity Expansion (DCE). You can add up to 12 disks to a pool at a time. When the DCE operation is initiated, part of the existing D-Piece elements that form the logical volumes is actually transferred to the new disks, thereby increasing the number of physical disks that host the logical volume and improving performance.
Another important advantage of the Dynamic Disk Pool is that instead of using dedicated isolated components to hot-swap failed disks, the pool itself has a built-in backup capacity on each disk included in the pool in case of possible failures. This simplifies management, as you no longer need to plan or monitor individual dedicated hot-swap drives. At the same time, recovery time is significantly reduced (two to three times when using large disks), and the performance of logical volumes during recovery is practically not reduced compared to traditional dedicated hot-swap disks.
In addition, given that virtual disks in Dynamic Disk Pools can contain a very large number of disks compared to traditional RAID 5 or RAID 6 disk groups with up to 30 disks per group, then in environments with mixed or non-parallel loads, you can achieve benefits in terms of performance, since all pool resources are available to all logical volumes at once.
In addition to the use of DDP, due to the possibility of combining SSDs and traditional hard drives in one solution, it became possible to implement caching on SSDs. SSD caching allows the PowerVault MD3 array to use solid-state drives as an extended read cache, thereby improving application performance with a large number of arbitrary read operations, for example for file servers, web servers, databases, etc. The
solid state drive cache is optional and is used along with the core to improve performance. When creating a cache on an SSD, it is divided into two internal virtual RAID 0 disks (one disk per controller). This cache volume is not available for normal data storage.
In full duplex mode, each controller uses a separate virtual disk, gaining access to only half of the SSD space, even if the controller fails or enters maintenance mode. In simplex mode, one controller manages the entire SSD space and uses both virtual disks. The maximum allowable cache size on a solid-state drive in PowerVault MD3 arrays is 4 TB.
SSD caching provides benefits for workloads with the following features:
- Read performance is limited by the speed of I / O operations for hard drives.
- High percentage of read operations relative to write operations.
- The size of the working data set is smaller than the cache size on the solid state drive.
SSD cash is recommended when a balanced, cost-effective approach is required using a combination of hard drives and cache on a solid state drive, and the cost of dedicated SSD drives for individual logical volumes is prohibitive.
Another feature of PowerVault MD3 storage is synchronous and asynchronous remote replication. It allows you to effectively replicate data, protecting information from system crashes and power failures. This feature duplicates data between PowerVault MD3 storage systems to ensure availability in the event of a failure, as well as for other business data migration purposes.
In the synchronous model, one array stores the primary virtual disk, and the second, remote, stores the secondary virtual disk of the replication pair. After establishing a connection between the primary and secondary arrays, the primary array processes the server-initiated write operations on the secondary virtual disk, sending write requests to the secondary array, while simultaneously performing these operations on the local array. The primary array returns to the server system a successful result only after both arrays have completed the write operation.
The synchronous model has several advantages, the most important of which is the ability to recover application data from a secondary array, including the latest updates. If a critical failure occurs on the primary array, then application data can be restored from the secondary array, then only write requests that were executed during the failure will disappear. Since the application on the side of the primary array expects a positive response about the completion of the current requests, their loss will not cause discrepancies in the data in the event of a server failure or reboot during their execution. Applications that have even the most basic recovery capabilities, in such cases, are able to restore data integrity on their own. Consequently, Using a synchronous remote replication model allows applications to realize fault tolerance. The disadvantages of synchronous replication include high requirements for communication channels.
Asynchronous remote replication is a great choice for those who need protection against failures, but who do not have the means to implement synchronous replication models. It is assumed that such consumers do not need an absolutely reliable recovery plan until the failure, they are more interested in the cost-effectiveness of the solution. In other words, asynchronous replication allows the consumer to use the slower and less expensive communication channels between the primary and secondary arrays if certain delays are allowed, which are not present when using the synchronous replication model. Asynchronous replication model allows you to adjust the duration of the delay depending on the needs of the client in accordance with the capabilities and characteristics of the communication channel. One of the main benefits of asynchronous remote replication is that that the performance of the application in the main array will not be related to the communication channel. Of course, channel performance will affect latency, but it will not directly affect application performance.
Asynchronous remote replication is designed to work with relatively slow and low-speed communication channels (for example, TCP / IP channels for iSCSI). The main mode of its operation includes customizable periods of divergence, during which no attempts are made to transfer data from the primary array to the secondary one. Instead, using a dirty card allows the array to track the changed portions of the primary virtual disk for subsequent selective transfer to the secondary array. After the time period specified in the settings has expired, the primary array starts transferring the corresponding data to the secondary array. This approach provides several advantages, each of which is especially important when using slow, global communication channels.
- Performance . During the discrepancy interval, no attempt is made to transmit changes through the communication channel. Accordingly, the performance on the primary array does not depend on the characteristics of the communication channel. The only reason that affects performance is the relatively low latency associated with tracking changed blocks. This is done by updating the dirty map on the primary array. During synchronization using the image at a certain point in time, it allows replication without significant dependence of the input-output speed on the primary virtual disk on the speed and delays of the communication channel.
- Efficiency. Since during the accumulation of changes during the divergence period, many updates can occur in the same blocks, asynchronous remote replication avoids the transmission of intermediate results. In fact, only the contents of the block existing at the time of synchronization are transmitted through the communication channel.
- Reliability . Much attention is paid to ensuring the validity of the contents of the block on the primary virtual disk and the possibility of its recovery before starting synchronization with the secondary array. In particular, asynchronous remote replication checks the contents of the block selected on the primary virtual disk for synchronization for compliance with the recording order in the primary array. As a result, the user can be sure that the data images on the secondary array will be suitable for recovery if it is necessary to make the secondary virtual disk active.
- Simplicity . If the secondary virtual disk becomes active due to the failure of the primary, there are situations when the functionality of the primary array can subsequently be restored. This can happen if the problem or failure on the primary side was caused by a power or communication failure, but it was later fixed. In such cases, asynchronous remote replication supports optimized reverse synchronization with the transfer to the original primary side of only differences in data. In other words, only those blocks that have changed since the transition of the secondary virtual disk to the active state will be transferred to the primary virtual disk. Therefore, it is possible to restore the original primary array in its role without resource-intensive full synchronization.
- Distance . Asynchronous remote replication supports FC and iSCSI networks. The iSCSI interface uses standard IP networks to replicate data over much longer distances than typical FCs. In addition, asynchronous remote replication does not require a dedicated server port for replication I / O; they are performed through the same iSCSI ports as the server I / O.
Asynchronous Remote Replication for Dell Storage Systems PowerVault MD3 provides a valuable solution for customers who need to provide fault tolerance to protect critical business data without the significant financial investment that requires synchronous remote replication. Asynchronous remote replication can be used on relatively slow communication channels, allowing the user to provide a reasonable compromise between RPO (Return Point Objective) and the cost of communication lines. Therefore, asynchronous remote replication allows you to transmit data over long distances via TCP / IP networks, providing protection against disasters, but without significantly affecting the performance of applications on the primary node.
In addition, using efficient MD Storage Manager software provides simple, easy-to-use, task-based configuration and monitoring tools. Add to this the VMware vStorage APIs for array integration (VAAI, VASA support), support for operating systems WS2008, Windows 2012, RHEL 5.8, 6.2, SLES 10.4, 11.2, Citrix XenServer, VMware ESXi 5.0, 5.1, MS Hyper-V 2008, work with thin volumes and you get a reliable system for using the most modern and most resource-intensive applications. Based on the requirements for the system, you can use the 90-day trial license to use the following features:
- work with snapshots and snapclones;
- productive configurations (High Performance Tier) - use of SSD cache on disks;
- using synchronous / asynchronous replication (between two arrays).
Activation is done through a graphical interface (MD Storage Manager) and automatically turns off after 90 days of use. At the same time, working data is not lost, but becomes inaccessible until the license is purchased.
Thus, PowerVault MD3 storage systems can find their application in a wide range of tasks. From general-class tasks (consolidation of business applications, virtualization projects, archive storage systems, etc.), to high-performance computing applications (computing clusters, video surveillance, digital content distribution).
For additional information on the configuration, capabilities or ordering conditions for the PowerVault MD3 family of storage systems, you can contact Marat Rakayev , the Dell Moscow office.
