In this article, we'll look into the future of SCSI and look at some of the advantages and disadvantages of SCSI, SAS, and SATA interfaces.
In fact, the issue is a bit more complex than just replacing SCSI with SATA and SAS. Traditional parallel SCSI is a tried and tested interface that has been around for a long time. Currently, SCSI offers a very fast data transfer rate of 320 Megabytes per second (Mbps) using the modern Ultra320 SCSI interface. In addition, SCSI offers big choice features, including Command-Tag Queuing (a method of optimizing I / O commands to increase performance). SCSI hard drives are reliable; in a short distance, you can create a daisy chain of 15 devices connected to a SCSI link. These features make SCSI an excellent choice for high performance desktops and workstations, up to and including enterprise servers, to this day.
SAS hard drives use the SCSI command set and have the same reliability and performance as SCSI drives, but use a serial version of the SCSI interface at 300 Mbps. Although slightly slower than 320 Mbps SCSI, the SAS interface is capable of supporting up to 128 devices over longer distances than the Ultra320 and can expand to 16,000 devices per channel. SAS hard drives offer the same reliability and rotation speeds (10000-15000) as SCSI drives.
SATA drives are a little different. Where SCSI and SAS drives focus on performance and reliability, SATA drives trade them off in favor of massive capacity increases and cost reductions. For example, a SATA drive in this moment reached a capacity of 1 terabyte (TB). SATA is used where maximum capacity is needed, for example, for Reserve copy data or archiving. SATA now offers point-to-point connections at speeds up to 300 Mbps, and easily outperforms the traditional parallel ATA interface at 150 Mbps.
So what will happen to SCSI? It works great. The problem with traditional SCSI is that it's just coming to the end of its useful life. Parallel SCSI at 320 Mb/s will not run much faster on current SCSI cable lengths. In comparison, SATA drives will reach 600 Mb/s in the near future, SAS have plans to reach 1200 Mb/s. SATA drives can also work with the SAS interface, so these drives can be used simultaneously in some storage systems. The potential for increased scalability and data transfer performance far exceeds that of SCSI. But SCSI isn't going away anytime soon. We will see SCSI in small and medium servers for a few more years. As hardware upgrades, SCSI will be systematically replaced by SAS/SATA drives to get faster and more convenient connections.
SAS interface.
The SAS or Serial Attached SCSI interface provides connectivity over a physical interface, similar to SATA, devices, command set-driven SCSI. Possessing backward compatible with SATA, it makes it possible to connect any devices controlled by the SCSI command set via this interface - not only hard drives, but also scanners, printers, etc. Compared to SATA, SAS provides a more developed topology, allowing parallel connection of one device over two or more channels. Bus expanders are also supported, allowing you to connect multiple SAS devices to a single port.
The SAS protocol is developed and maintained by the T10 committee. SAS was designed to communicate with devices such as hard drives, storage optical discs and the like. SAS uses a serial interface to work with direct-attached drives, compatible with the SATA interface. Although SAS uses a serial interface as opposed to the parallel interface used by traditional SCSI, SCSI commands are still used to control SAS devices. Commands (Fig. 1) sent to the SCSI device are a sequence of bytes of a certain structure (command descriptor blocks).
Rice. one.
Some commands are accompanied by an additional "parameter block" that follows the command descriptor block, but is already passed as "data".
A typical SAS interface system consists of the following components:
1) Initiators. An initiator is a device that originates service requests for target devices and receives acknowledgments as the requests are executed.
2) Target Devices. The target device contains logical blocks and target ports that receive service requests and execute them; after the processing of the request is completed, a confirmation of the request is sent to the initiator of the request. The target device can be either a separate hard drive, and the whole disk array.
3) Data delivery subsystem. It is part of the I / O system that transfers data between initiators and target devices. Typically, the data delivery subsystem consists of cables that connect the initiator and the target device. Additionally, in addition to cables, the data delivery subsystem may include SAS extenders.
3.1) Expanders. SAS extenders are devices that are part of the data delivery subsystem and make it possible to facilitate data transfers between SAS devices, for example, allowing you to connect several target SAS devices to one port of the initiator. Connecting through an extender is completely transparent to target devices.
SAS supports connecting SATA devices. SAS uses a serial protocol to transfer data between multiple devices and thus uses fewer signal lines. SAS uses SCSI commands to manage and communicate with target devices. The SAS interface uses point-to-point connections - each device is connected to the controller by a dedicated channel. Unlike SCSI, SAS does not require the user to terminate the bus. The SCSI interface uses a common bus - all devices are connected to the same bus, and only one device can work with the controller at a time. In SCSI, the speed of information transfer on different lines that make up a parallel interface can vary. The SAS interface does not have this shortcoming. SAS supports a very large number of devices, while SCSI supports 8, 16, or 32 devices on the bus. SAS supports high data rates (1.5, 3.0, or 6.0 Gbps). Such a speed can be achieved by transferring information on each connection, while on the SCSI bus, the bus bandwidth is divided between all devices connected to it.
SATA uses the ATA command set and supports hard drives and optical drives, while SAS supports a wider range of devices, including hard drives, scanners, and printers. SATA devices are identified by the port number of the SATA interface controller, while SAS devices are identified by their WWN (World Wide Name) identifiers. SATA devices (version 1) did not support command queues, while SAS devices support tagged command queues. SATA devices since version 2 support Native Command Queuing (NCQ).
SAS hardware communicates with target devices on several independent lines, which increases the fault tolerance of the system (the SATA interface does not have this capability). At the same time, the interface SATA versions 2 uses port duplicators to achieve a similar capability.
SATA is predominantly used in non-critical applications such as home computers. The SAS interface, due to its reliability, can be used in mission-critical servers. Error detection and error handling is much better defined in SAS than in SATA. SAS is considered a superset of SATA, and does not compete with it.
SAS connectors are much smaller than traditional parallel SCSI connectors, allowing SAS connectors to be used to connect 2.5" compact drives. SAS supports data transfer rates from 3 Gb/s to 10 Gb/s. There are several options for SAS connectors:
SFF 8482 is a variant compatible with the SATA interface connector;
SFF 8484 - internal connector with dense packing of contacts; allows you to connect up to 4 devices;
SFF 8470 - densely packed connector for connection external devices; allows you to connect up to 4 devices;
SFF 8087 - reduced Molex iPASS connector, contains a connector for connecting up to 4 internal devices; supports 10 Gbps;
SFF 8088 - reduced Molex iPASS connector, contains a connector for connecting up to 4 external devices; supports 10 Gbps speed.
The SFF 8482 connector allows you to connect SATA devices to SAS controllers, which eliminates the need to install an additional SATA controller just because, for example, you need to connect a device for recording DVD discs. Conversely, SAS devices cannot connect to the SATA interface, and a connector is installed on them to prevent them from connecting to the SATA interface.
What is SAS, background It's time to face the obvious: the SCSI standard, even in the most modern implementations like Ultra320 SCSI, has exhausted its capabilities. At the very least, further scaling of its performance, if theoretically possible, will be very expensive. The situation with this highly respected standard looks especially depressing against the backdrop of the rapid development of the entire computer technology and architectures and topologies of storage systems in particular.
Two key factors pushing manufacturers to improve hard drive interfaces are the growing performance of data storage systems with a large number of serviced transactions and the speed of retrieving data from large libraries. Of course, "a holy place is never empty", and the appearance of interfaces like optical FCAL or serial SATA to some extent made it possible to get rid of "bottlenecks" and diversify the list of storage system architectures. However, users accustomed to the possibilities of SCSI still remain fans of this standard. Moreover, a lot of money has been invested in its development.
These are the prerequisites for the birth of a new industrial standard, called serial-attached SCSI - Wikiwand Serial-Attached SCSI, or simply SAS.
To be fair, it should be noted that new standard did not appear suddenly and immediately: the official announcement of SAS technology, which took place on January 28, 2004, was preceded by serious work by the development team from different companies and industry groups - SCSI Trade Association (STA) and International Committee for Information Technology Standards (INCITS), under the auspices of the American National Standards Institute (ANSI). The new standard was first talked about in December 2001, when the board of directors of the SCSI Trade Association (STA) voted to define Serial Attached SCSI specifications. Further, on May 2, 2002, the development of the standard was transferred to the T10 committee at INCITS (InterNational Committee for Information Technology Standards), created specifically to support, develop and promote SAS, and the first draft SAS specifications were published in mid-2003.
So, the most important thing to rely on when trying to formulate a definition of the SAS standard: Serial-Attached SCSI is a logical and natural serial extension of the SCSI parallel interface technology used to connect peripherals to computers.
From this, for starters, and push off.
Purpose of the SAS
To determine the purpose of the SAS standard and its place among modern peripheral interfaces, let's turn to the wording set forth in the "FAQ on Serial Attached SCSI" on the T10 website.The Serial Attached SCSI interface is the product of a logical evolution modern interfaces and designed for use in industrial data centers. The SAS standard relies on the electrical and physical characteristics of the Serial ATA interface to provide scalability, performance, reliability, and data manageability in servers and storage subsystems. The architectural similarity with SATA does not prevent SAS from having the most sought-after features of SCSI, while at the same time getting rid of its disadvantages: large connectors, short lengths of connecting cables, limited performance and addressing.
In a broad sense, SAS is a kind of full-duplex SATA with support for two ports, greater addressing capabilities, enhanced reliability, performance, and logical compatibility with SCSI. The Serial ATA interface, on the other hand, can be viewed as a simplified subset of Serial Attached SCSI for operation in simple systems without critical reliability and performance requirements. This does not mean at all that Serial Attached SCSI devices cannot be used in ordinary workstations and desktop PCs, only the presence of an appropriate host adapter is required.
In fact, Serial Attached SCSI is SCSI, but not with the usual parallel, but with a point-to-point (point-to-point) serial architecture, with a direct connection of the controller to the drives. SAS supports up to 128 drives various types and sizes connected together by thinner and longer (than in the case of SCSI) cables. While the SCSI interface "pushes" data through its wires at a rate of about 20 MB / s, and half-duplex first generation SATA - 1.5 GB / s in one direction per unit of time, a full-duplex SAS signaling serial interface with support for "hot" connection in The current implementation provides data exchange at speeds up to 3.0 Gb / s per port.
The key difference between SAS and SCSI is the ability to connect SAS drives to two different ports simultaneously, each representing a different SAS domain. You can imagine how significantly this affects the reliability of data storage and system fault tolerance. In addition, the "switching" nature of the SAS architecture allows in theory to connect thousands of drives "cascaded" (up to 16384 drives without performance degradation!), which makes the scalability of such systems theoretically unlimited. The main differences between SCSI and SAS technologies are shown in the table below.
SAS Connector and Cable Specifications
One of key features The SAS interface during its development determined the possibility of a significant increase in the speed of data exchange. The next-generation SAS specifications currently under development include data transfer rates up to 6.0 GB/s with full compatibility with the first generation of SAS devices. The next generation after this has not been seriously considered yet, but they are talking about the possibility of achieving a data exchange rate of up to 12 GB / s.
When developing connectors for SAS devices, a promising increase in data exchange speed was laid, and at the same time, the miniaturization experience seen in the SATA specifications was taken into account. The specificity of the connector lies in the placement of the second data port, because each of the ports of the SAS device is located in different domains and serves to organize independent paths from one SAS device to another to ensure trouble-free operation. If one of the drives in the chain fails, this in no way affects the operation of other devices. Thus, the design of the connector for peripherals with the SAS interface was born, which in fact has an architectural similarity with 68-pin connectors for drives with the classic parallel SCSI or SCA-2 interface, but at the same time, by analogy with SATA, which supports "hot plugging". and reliable contact.
The SAS cable system is much more compact than parallel ATA and SCSI cabling, resulting in less clutter and better airflow for the components inside the system case. The typical length of SAS interface cables for applications such as workstations is less than 1 m, the maximum length of such a cable can be up to 8 m. Theoretically, this is comparable to the length of a cable for a SCSI interface, since some modern devices allow a connection between the host controller and SCSI - periphery at a distance of more than 8 m. However, in case of need, the distance between SAS devices can be significantly increased due to the so-called SAS expanders - a kind of "pipeline pumping station".
It is interesting to note that when developing the SAS specifications, the working group immediately took into account the need to define the parameters of connectors and cables, not only for internal, but also for external connections, similar to modern SCSI options like "server - JBOD system". For the SATA interface, the adoption of such specifications was postponed "until later", and, as a result, the development of External SATA is still not finished.
As for external SAS connections, the proposal of Infiniband was taken as the basis, where external connectors and cabling are designed for 4 devices and at the same time provide the performance of the first generation of external SAS connections at 1.2 Gb / s in each direction, i.e. up to 2400 MB/s full duplex! Agree, more than impressive for the external interface.
SAS system topology
The use of point-to-point class configurations makes it possible to obtain high throughput, however, the reverse side of the coin is the organization of a specific topology, where the interaction of initiating (host) devices and peripherals implies support for more than two devices "in a bundle". When the SAS standard was developed, the specification immediately included the existence of inexpensive expanders that allow you to create systems with more than one initiating hosts, with support for more than one peripheral device.Another important goal that the developers of the new standard set for themselves was to get away from the classic SCSI limitation, which implies no more than 16 devices in one chain. As a result, each SAS system, when using the appropriate number of expanders, is able to support addressing up to 16256 devices in a single SAS domain. Be sure to note the configuration flexibility of SAS expanders: their specifications imply the creation of heterogeneous systems, where both SAS and SATA devices can coexist as peripheral drives. Agree, it is very convenient, especially when forming budget systems storage of data or devices with future scaling.
Illustration for the principle of organizing a SAS domain
maximum capacity
Pay attention to the illustration above: the dark green module in the center is the same expander-switch (fanout expander). Such a "switching" expander can be present in a single SAS domain and unite up to 128 SAS devices. However, SAS devices should not be understood exclusively hard drives, since here we mean any possible combination of the so-called "edge expanders" (edge expanders, light green modules), initiating devices and the drives themselves. Peripheral expanders, in turn, can also support up to 128 SAS devices, however, no more than one additional expander can be connected to them. The blue modules in the diagram are initiators (hosts), and the brown cylinders are SAS or SATA drives.
SAS protocols
The creation of a new topology and new interfaces has led to an entirely new definition of how to address all possible ports in a SAS domain. With parallel SCSI, of course, everything is simpler, since the addressing of all devices in the domain is predetermined at the hardware level.As a result working group On the development of the SAS protocol, it was decided to choose globally unique 64-bit names - WWN (WorldWide Name) as identifiers for all types of SAS devices. Again, nothing new under the sun, it is this addressing that has long been used when naming Fiber Channel devices.
Thus, at the moment of power-up, all devices united in a single SAS space exchange their WWNs with each other, and only after that the set of SAS devices becomes a "meaningful" SAS system. Adding a new device to the SAS system (in this case, adding it means just "hot plugging") or removing it from the system leads to a notification that notifies all initiators of the event and allows you to adjust the system to a new configuration. The expanders, in turn, are responsible for issuing a WWN to all SATA devices in the system, both when it is turned on and when a new device is "hot" plugged in. Upon completion of the system initialization process, SATA devices communicate using SATA protocols, for SAS devices, the SAS protocol is used, described in other SCSI standards such as SPI (SCSI Parallel Interface).
Further, everything is simpler: the exchange of commands, data, statuses and other information between SAS devices is carried out in packets, the specifications of which are very similar to the characteristics of packets for exchanging information when working with parallel SCSI or Fiber Channel devices. The format of SAS data packets, called "frames", is especially similar to the Fiber Channel specifications: each of them consists of command descriptor blocks - CDB (command descriptor block) and other SCSI constructs defined by other SCSI standards, such as SCSI Primary Command Set or SCSI block command. Here's another benefit of the SAS standard: using a SCSI-like protocol and architecture allows you to combine SAS designs with other storage and processing systems with an Infiniband, iSCSI or Fiber Channel architecture, which, in fact, are also SCSI objects.
The SAS protocol contains four traditional layers: physical (phy layer), communication (link layer), port layer (port layer) and transport layer (transport layer). The combination of four layers in each SAS port means that programs and drivers used to work with parallel SCSI ports can be used equally well for servicing SAS ports, with only minor modification.
SAS architecture
Application layers, including drivers and applications themselves, create specific tasks for the transport layer, which, in turn, encapsulates commands, data, statuses, etc. in SAS frames and delegates their transmission to the port layer. Of course, the transport layer is also responsible for receiving SAS frames from the port layer, disassembling the received frames, and passing content to the application layer.
The SAS port layer is responsible for exchanging data packets with the communication layer (link layer) in the order of establishing connections, as well as for choosing the physical layer with which the packets will be transmitted simultaneously to several devices. The SAS physical layer refers to the corresponding hardware environment - transceivers and encoding modules that connect to the SAS physical interface and send signals over wired circuits.
By the way, let me remind you that at the physical level, connections in the case of a serial SAS interface are full-duplex differential pairs of circuits that can also be combined to increase performance (well, just like PCI Express) to "wide" ports. Accordingly, each device can have more than one port, and each of them can be configured as "narrow" or "wide". Host and expander interfaces can be made up of multiple ports, with each host's address available to everyone. peripheral device, and the throughput is summed up. The organization of multiple data paths due to the presence of "wide" ports implies parallel execution of commands and a corresponding reduction in the time spent waiting in line.
Conclusion
The presented material is only a brief introduction to the principles of constructing the architecture of the SAS interface and the features of the implementation of this standard. A more detailed consideration of the interface specifications will most likely require the release of a whole series of articles on this topic. It is possible that this will be the case, fortunately, the beginning of the mass implementation of the interface is just around the corner, and the number of applied questions on the implementation of SAS systems will only increase over time.The main definition of SAS, which, in my opinion, should not be forgotten - the new Serial Attached SCSI serial interface was designed for the needs of a wide range of enterprise-level storage systems, however, it is still a "close action" interface and in no way designed to replace any network interfaces, there is no need to "buy in" for a similar implementation of the "point-to-point" architecture.
For all its "sharpening" for working in large and almost infinitely scalable storage systems, the Serial Attached SCSI interface implies full compatibility with relatively inexpensive Serial ATA drives, which allows you to design quite affordable systems even at the scale of small enterprises. At the same time, support for 2-port Serial Attached SCSI drives allows for performance levels that are beyond the reach of today's SCSI drive systems.
For those who are ready to plunge into the study of the features of Serial Attached SCSI on their own, we conclude with a list of sites where educational and standard documents are located.
Adaptec website resources
Maxtor website resources
Seagate website resources
T10 :
Serial Attached SCSI -
SCSI Architecture Model-3 (SAM-3)
SCSI Primary Commands-3 (SPC-3)
SCSI Block Commands-2 (SBC-2)
SCSI Stream Commands-2 (SSC-2)
SCSI Enclosure Services-2 (SES-2)
SAS connector specifications:
SFF 8482 (internal backplane/drive)
SFF 8470 (external 4-wide)
SFF 8223, 8224, 8225 (2.5", 3.5", 5.25" form factors)
SFF 8484 (internal 4-wide)
Serial ATA specifications:
Serial ATA II: Extensions to Serial ATA 1.0
Serial ATA II: Port Multiplier
Serial ATA II: Port Selector
Serial ATA II: Cables and Connectors Volume 1
Additional resources:
International Committee for Information Technology Standards
T11 (Fiber Channel standards)
SCSI Trade Association
SNIA (Storage Networking Industry Association)
High-performance server drives for mission-critical tasks are rarely seen by IT publications. No wonder, because we are more focused on the mass buyer than on system administrators and suppliers server hardware. Meanwhile, testing server HDDs is even more important than testing desktop ones, for several reasons. Firstly, due to the higher cost of drives and the higher sensitivity of server tasks to performance. After the mass distribution of solid-state drives, the differences between desktop drives have ceased to be of great importance, and in a server, replacing an HDD with an SSD is far from always advisable. The following circumstance follows from the first: HDD for a desktop or home NAS can be chosen according to the basic technical specifications(volume, spindle speed, plate capacity). In the case of a server HDD, a lot depends on the optimization of the firmware, which manifests itself in a complex load and, accordingly, requires special tests to capture these features. Finally, at large scales, such a parameter as the ratio of performance to power consumption of the drive comes into play.
Over the past few years, the choice hard drives corporate purpose has definitely become easier. Models with Fiber Channel and SCSI interfaces have ceased to be produced. Drives are divided into two classes: models in the 3.5-inch form factor are limited to 7200 rpm, have a SAS or SATA interface - to choose from and are designed to store "cold" data (nearline storage). Drives with a speed of 10,000-15,000 rpm use the SAS interface and, for the most part, have moved to the 2.5-inch form factor (SFF - Small Form Factor), which allows you to increase the number of spindles per unit in the rack. Only HGST still has 15K-class drives in 3.5-inch form factor with Fiber Channel ports.
We are already constantly paying attention to nearline drives in a SATA configuration, but the test of SAS / SCSI drives is published for the first time on 3DNews.
⇡ Test participants
The following devices took part in the comparison:
- HGST Ultrastar C10K1800 1.8TB (HUC101818CS4200);
- HGST Ultrastar C15K600 600 GB (HUC156060CSS200);
- Seagate Savvio 10K.6 900 GB (ST900MP0006);
- Seagate Enterprise Performance 10K HDD v7 1.2TB (ST1200MM0017);
- Seagate Enterprise Performance 15K HDD v5 600 GB (ST600MP0035);
- Toshiba AL13SEB 900 GB (AL13SEB900);
- Toshiba AL13SXB 600 GB (AL13SXB600N);
- WD VelociRaptor 1TB (WD1000DHTZ).
Unlike desktop and NAS hard drives, SAS drives are not so different from each other. All participants:
a) are available in a 2.5-inch form factor with a thickness of 15 mm;
b) have two SAS ports to improve fault tolerance;
c) prepared for 24/7 operation in a telecommunications rack;
d) allow the user to configure the sector size for recording additional metadata;
e) are characterized by the same reliability indicators (MTBF, number of head parking cycles);
e) are sold with a five-year manufacturer's warranty.
For testing, models of the maximum volume in the corresponding lines were selected. The products of all companies that produce HDDs today are presented, with one exception. We have exhausted all the possibilities to get a WD Xe drive for a test (except just to buy it for a lot of money), and recently this brand has completely disappeared from the corporate website western digital Apparently out of production. As a result, of all drives with a spindle speed of 10-15 thousand rpm, WD has only VelociRaptor - in fact, a derivative of WD Xe, but with a SATA interface. In order for WD to be at least somehow represented in the review, we included the VelociRaptor in the number of participants. Of course, it cannot be considered a 100% replacement for SAS drives, but a lot of servers run on SATA drives, so VelociRaptor can also be used. In addition, if you look at the other side, any of the drives for SAS can be used in a workstation with the appropriate HBA (Host Bus Adapter) instead of VelociRaptor, which also justifies the participation of this drive in today's test.
| Manufacturer | HGST | HGST | Seagate | Seagate | Seagate | Toshiba | Toshiba | western digital |
| Series | Ultrastar C10K1800 | Ultrastar C15K600 | Savvio 10K.6 | Enterprise Performance 10K HDD v7 | Seagate Enterprise Performance 15K HDD v5 | AL13SEB | AL13SXB | VelociRaptor |
| Model number | HUC101818CS4200 | HUC156060CSS200 | ST900MM0006 | ST1200MM0017 | ST600MP0035 | AL13SEB900 | AL13SXB600N | WD1000CHTZ/WD1000DHTZ |
| Form Factor | 2.5 inches | 2.5 inches | 2.5 inches | 2.5 inches | 2.5 inches | 2.5 inches | 2.5 inches | 3.5/2.5 inch |
| Interface | SAS 12Gb/s | SAS 12Gb/s | SAS 6Gb/s | SAS 6Gb/s | SAS 12Gb/s | SAS 6Gb/s | SAS 6Gb/s | SATA 6Gb/s |
| dual-port | Yes | Yes | Yes | Yes | Yes | Yes | Yes | Not |
| Capacity, GB | 1 800 | 600 | 900 | 1 200 | 600 | 900 | 600 | 1000 |
| Configuration | ||||||||
| Spindle speed, rpm | 10 520 | 15 030 | 10 000 | 10 000 | 15 000 | 10 500 | 15 000 | 10 000 |
| Data recording density, GB/platter | 450 | 200 | 300 | 300 | 200 | 240 | ND | 334 |
| Number of plates/heads | 4/8 | 3/6 | 3/6 | 4/8 | 3/6 | 4/8 | ND | 3/6 |
| Buffer size, MB | 128 | 128 | 64 | 64 | 128 | 64 | 64 | 64 |
| Sector size, bytes | 4096-4224 | 512-528 | 512-528 | 512-528 | 4096-4224 | 512-528 | 512-528 | 512 |
| Performance | ||||||||
| Max. sustained sequential read speed, MB/s | 247 | 250 | 195 | 195 | 246 | 195 | 228 | 200 |
| Max. sustained sequential write speed, MB/s | 247 | 250 | 195 | 195 | 246 | 195 | 228 | 200 |
| Burst rate, read/write, MB/s | 261 | 267 | ||||||
| Internal data transfer rate, MB/s | 1307-2859 | 1762-3197 | 1440-2350 | 1440-2350 | ND | ND | ND | ND |
| Average seek time: read/write, ms | 3,7/4,4 | 2,9/3,1 | ND | ND | ND | 3,7/4,1 | 2,7/2,95 | ND |
| Track-to-track seek time: read/write, ms | ND | ND | ND | ND | ND | 0,2/22 | ND | ND |
| Full stroke seek time: read/write, ms | 7,3/7,8 | 7,3/7,7 | ND | ND | ND | ND | ND | ND |
| Reliability | ||||||||
| MTBF (mean time between failures), h | 2 000 000 | 2 000 000 | 2 000 000 | 2 000 000 | 2 000 000 | 2 000 000 | 2 000 000 | 1 400 000 |
| AFR (annualized failure rate), % | ND | 0,44 | 0,44 | 0,44 | 0,44 | ND | 0,44 | ND |
| Number of head parking cycles | 600 000 | 600 000 | ND | ND | ND | ND | 600 000 | 600 000 |
| physical characteristics | ||||||||
| Power consumption: idle / read-write, W | 5,4/7,6 | 5,8/7,5 | 3,9/7,8 | 4,6/8,1 | 5,3/8,7 | 3.9/ND | 5,0/9,0 | 4,2/5,8 |
| Typical noise level: idle/searching | 34/38 dBA | 32/38 dBA | 30 dBA / ND | 31 dBA / ND | 32.5/33.5 dBA | 30 dBA / ND | 33 dBA / ND | 30/37 dBA |
| Maximum temperature, °C: disk on / disk off | 55/70 | 55/70 | 60/70 | 60/70 | 55/70 | 55/70 | 55/70 | 55/70 |
| Shock resistance: drive enabled (read) / drive disabled | 30 g (2 ms) - recording / 300 g (2 ms) | 25 g (2 ms) / 400 g (2 ms) | 25 g (2 ms) / 400 g (2 ms) | 25 g (2 ms) / 400 g (2 ms) | 100 g (1 ms) / 400 g (2 ms) | 100 g (1 ms) / 400 g (2 ms) | 30 g (2 ms) / 300 g (2 ms) | |
| Dimensions: L × H × D, mm | 101×70×15 | 100×70×15 | 101×70×15 | 101×70×15 | 101×70×15 | 101×70×15 | 101×70×15 | 101 x 70 x 15/ 147 x 102 x 26 |
| Weight, g | 220 | 219 | 212 | 204 | 230 | 240 | 230 | 230/500 |
| Warranty period, years | 5 | 5 | 5 | 5 | 5 | 5 | 5 | 5 |
| Average retail price, rub.* | 161 000 | 36 000 | 20 000 | 26 900 | 49 600 | 17 800 | 24 100 | 14 000 / 12 600 |
⇡ Description of test participants
HGST Ultrastar C10K1800 1.8TB (HUC101818CS4200)
This is the largest drive in HGST's latest 10K lineup. The Ultrastar C10K1800 series is notable in several respects. Models ending in S420x achieve 450 GB per platter thanks to their high recording density using 4K sector formatting (native or 512-byte sector emulation). Therefore, the disk can hold up to 1.8 TB, and the sequential read / write speed has reached the level of the HDD class of 15 thousand rpm.
The rest of the line consists of disks with a markup of 512-528 bytes, with less outstanding speed and up to 1.2 TB.
All models in the C10K1800 line have a so-called media cache. In several places on the surface of the plates, areas serving as a non-volatile cache are highlighted. Instead of carrying data to the requested sector, the write head of the disk flushes it to the nearest cache area, and when the disk is idle, it is moved to the right place.
Incidentally, this is the most expensive disc in the test, fantastically expensive - an average of 161,000 rubles in Moscow online stores. And in America, by the way, it's much cheaper - $800 at newegg.com.
|
|
|
|
|
HGST Ultrastar C10K1800 1.8TB (HUC101818CS4200) |
||
HGST Ultrastar C15K600 600 GB (HUC156060CSS200)
The only 15K RPM 2.5" drive line in the HGST range. Ultrastar C15K600 drives simultaneously have the highest sequential read/write speed and low latency at the same time. Physical formatting of plates is performed in sectors of 512-528 or 4096-4224 bytes (with native access or 512 bytes emulation). The test involves the most capacious model in the line - 600 GB with 4 KB sectors.
|
|
|
|
|
HGST Ultrastar C15K600 600 GB (HUC156060CSS200) |
||
Seagate Savvio 10K.6 900 GB (ST900MP0006)
These are rather old drives - the generation before last compared to the current Enterprise Performance 10K line from Seagate. Therefore, the performance of the Savvio 10K.6 is no longer class-leading. The plates were formatted in sectors of 512-528 bytes. However, these disks are still on sale, they have a good volume (up to 900 GB) and are relatively inexpensive.
|
|
|
|
|
Seagate Savvio 10K.6 900 GB (ST900MP0006) |
||
Seagate Enterprise Performance 10K HDD v7 1.2TB (ST1200MM0017)
This series also formally became obsolete by the time the test was released, giving way to Enterprise Performance 10K HDD v8. These drives differ from Savvio 10K.6 only in increased capacity up to 1.2 TB, but this was achieved by increasing the number of platters, not the recording density, so there is no difference with the previous generation in terms of the declared performance. The model ST1200MM0017 participating in testing has built-in encryption.
|
|
|
|
|
Seagate Enterprise Performance 10K HDD 1.2TB (ST1200MM0007) |
||
Seagate Enterprise Performance 15K HDD v5 600 GB (ST600MP0035)
it current line Seagate drives with a spindle speed of 15 thousand rpm. Disks have sector markings of 512-528 or 4096-4224 bytes (natively or with 512 bytes emulation). The maximum capacity (600 GB) drive with 4-kilobyte sectors was tested.
|
|
|
|
|
Seagate Enterprise Performance 15K HDD 600 GB (ST600MP0035) |
||
Toshiba AL13SEB 900 GB (AL13SEB900)
According to the main characteristics, this is an analogue of Seagate Savvio 10K.6: 10,000 rpm, volume up to 900 GB, formatting by sectors of 512-528 bytes. Toshiba does not offer drives with built-in encryption in this series.
|
|
|
|
Toshiba AL13SXB 600 GB (AL13SXB600N)
In this series of 15,000 rpm discs, models with names like AL13SXB**0N are formatted with a sector size of 512-528 bytes. We took the oldest of them for testing. Models with names like AL13SXB**E* use 4K sectors and also support a 12Gb/s SAS interface. There is no built-in encryption in the entire AL13SXB series.
|
|
|
|
|
Toshiba 900 GB (AL13SEB900) |
||
WD VelociRaptor 1TB (WD1000CHTZ/WD1000DHTZ)
In terms of physical data, VelociRaptor differs little from its prototype - WD Xe: the same 10,000 rpm and almost the same linear performance. VelociRaptor uses Advanced Format partitioning (4 KB sectors), and the amount available to the user is higher than that of similar WD Xe (1 TB in the case of the older model).
Since this is a SATA drive, functionally it is not a complete analogue of SAS drives. In particular, two-port connectivity, sector size configuration, and built-in encryption can be forgotten. In addition, SAS drives are usually made more reliable, which is noticeable when comparing their claimed MTBF with that of the VelociRaptor. Still, in terms of performance, this drive can be considered as a server ten-thousander for the poor. There are versions of the "lizard" with a radiator-adapter to the form factor 3.5 inches (DHTZ), as well as "naked" versions of 2.5 inches (CHTZ).
|
|
|
|
|
WD VelociRaptor 1TB (WD1000DHTZ) |
||
⇡ Testing methodology
Isolated Performance Tests
Performed using Iometer 1.1.0. The volume and speed of data transfer is indicated in binary units (1 KB = 1024 bytes). Block boundaries are aligned with the 4 KB markup.
- Sequential read/write of 128 KB block data with a request queue depth of 256.
- Random read / write blocks from 512 bytes to 2 MB with a request queue depth of 256.
- Mixed read/write of 128 KB blocks with a request queue depth of 256. The share of read and write operations varies from 0 to 100% in 10% increments.
- Addiction bandwidth on the length of the command queue. Reading blocks of 4 KB is performed, the depth of the request queue varies from 1 to 256 in steps of a power of two. A similar test for writing blocks is not carried out, because. hard drives do not differ in this parameter.
- Steady response time. Random read/write of 512-byte blocks with a request queue depth of 1 is performed. The test continues for 10 minutes.
- Constancy of response time. Random reading/writing of blocks of 4 KB in size with a request queue depth of 256 is performed. For each segment of the test with a duration of 1 s, the average and maximum values of the response time are recorded, on the basis of which: a) the average values of both indicators are calculated; b) standard deviation of the average response time.
- Multi-threaded read/write. Four threads are created that perform sequential reading/writing of 64 KB blocks with a request queue depth of 1. The threads have access to non-overlapping 100 GB address spaces, which are located close to each other in the disk volume, starting from sector zero. The aggregate throughput of all streams is measured, as well as each of them individually.
Simulated Load Tests
Run in Iometer 1.1.0. The volume and speed of data transfer is indicated in binary units (1 KB = 1024 bytes). Block boundaries are aligned with the 4 KB markup. The command queue depth is 256.
| Block size | Share of all requests | Share reading | Random access share |
| Database | |||
| 8 KB | 100% | 67% | 100% |
| File server | |||
| 512 bytes | 10% | 80% | 100% |
| 1 KB | 5% | 80% | 100% |
| 2 KB | 5% | 80% | 100% |
| 4 KB | 60% | 80% | 100% |
| 8 KB | 2% | 80% | 100% |
| 16 KB | 4% | 80% | 100% |
| 32 KB | 4% | 80% | 100% |
| 64 KB | 10% | 80% | 100% |
| Work station | |||
| 8 KB | 100% | 80% | 80% |
| Web server | |||
| 512 bytes | 22% | 100% | 100% |
| 1 KB | 15% | 100% | 100% |
| 2 KB | 8% | 100% | 100% |
| 4 KB | 23% | 100% | 100% |
| 8 KB | 15% | 100% | 100% |
| 16 KB | 2% | 100% | 100% |
| 32 KB | 6% | 100% | 100% |
| 64 KB | 7% | 100% | 100% |
| 128 KB | 1% | 100% | 100% |
| 512 KB | 1% | 100% | 100% |
test bench
The drives were connected to the LSI SAS 9211-8i adapter, for which we express our gratitude to the Russian representative office of LSI.
⇡ Performance, basic tests
Sequential Read/Write
- Drives with a spindle speed of 15 thousand rpm rule the ball in the sequential read / write test. However, this group has its own leader - Seagate Enterprise Performance 15K HDD v5.
- Ultrastar C10K1800 is not inferior to 15K category drives due to high recording density.
- But the presented ten-thousanders differ little in terms of linear access speed.
Free reading
- 15-thousanders and in this discipline dominate over their low-speed counterparts.
- The spread of indicators within HDD categories with the same spindle speed is small. We can single out only the HGST Ultrastar C15K600 as the formal leader in its group and the VelociRaptor, which is clearly inferior to its counterparts.
Arbitrary entry
- The HGST Ultrastar C15K600 demonstrated tremendous performance, completely unattainable for competing devices.
- The two remaining 15K drives also have a big advantage over HDDs with lower spindle speeds.
- The 10Ks themselves make up a homogeneous group, with the exception of the Ultrastar C10K1800. It goes far beyond its class and is second only to the C15K600 drive from the same manufacturer. Here it is, the vaunted media cache, in action!
The results of the random write test turned out to be less predictable than in the previous test, since they are determined not only by the mechanics of the HDD, but also by the nature of the buffer usage.
Steady response time
- Although the load continues for 10 minutes, it may not completely fill the buffer on some drives, so the results for data writes do not reflect what this test is aimed at - the latency of the drive mechanics.
- On the contrary, when reading with a queue length of one instruction, the buffer is not a helper. As a result, the rivals lined up in accordance with the speed of rotation of the spindle (the higher it is, the faster the response time). No significant difference was found between devices of the same category.
⇡ Performance, advanced analysis
Mixed Read/Write
- The 15K drives are still on top, with the exception of the Ultrastar C15K600, which sank especially hard under mixed loads.
- Ultrastar C10K1800 has once again stood out among its peers. Of the other ten-thousanders, we note Toshiba AL13SEB. They are all about the same at 100 percent read or write, but the AL13SEB retains the best performance under mixed workloads.
Dependence of throughput on the length of the command queue
- All drives are able to benefit from a long queue of commands and reach peak throughput at 64 commands. Only the VelociRaptor is content with a queue of 32 teams.
Multi-threaded reading
- Most of the participants in the test evenly distribute resources among the four threads. Which, however, leads to low overall productivity.
- Toshiba AL13SEB and WD VelociRaptor, on the contrary, sacrifice one of the threads during multi-threaded reading, thereby increasing the data transfer rate in the others and the overall throughput.
Multi-threaded recording
- When writing to four streams, none of the disks is cunning: performance is evenly distributed between all streams.
- As you can see, not so much depends on the mechanics of the disk in this test. The 15K models from Seagate and Toshiba took the first places, but the Ultrastar 15K600 is an obvious outsider.
Response time consistency
- When reading data, all drives are characterized by a significant difference between the average and maximum response times. Only the VelociRaptor stands out with a better average to maximum response time ratio.
- When recording, the peak values of the response time are smoothed by the buffer and differ little from the average.
- The participants of the test differ most of all in the spread of write access times. The Ultrastar C10K1800 has the most consistent performance. Toshiba AL13SEB900, on the other hand, has a much higher standard deviation of access time.
Among server ten-thousanders, disks do not differ so much from each other, but formally, Seagate Savvio 10K.6 achieved the best results. VelociRaptor, on the other hand, always trails behind.
Most ten-thousanders are similar in their main aspects, but it is worth highlighting the HGST Ultrastar C10K1800 (HUC101818CS4200), which is inferior to more resourceful colleagues of the 15K class only in random read speed and at the same time has a record volume of 1.8 TB. However, these advantages did not affect the results of tests with emulated applications.
Seagate Savvio 10K.6 900 GB (ST900MP0006) and Seagate Enterprise Performance 10K HDD v7 1.2 TB (ST1200MM0007) deliver consistently high performance without surprises. Toshiba AL13SEB900 coped with the tests a little worse than other ten-thousanders.
The WD VelociRaptor 1TB (WD1000DHTZ) can be considered a "poor man's" high-performance HDD if the SAS protocol is not a mandatory item in the terms of reference. According to its characteristics, this is a typical 10K class disk, only in comparison with true server drives, the random read speed leaves much to be desired, which also manifested itself in "emulators".
SAS (Serial Attached SCSI)- serial computer interface designed to connect various devices data storage, for example, and tape drives. SAS is designed to replace the parallel SCSI interface and uses the same SCSI command set.
SAS is backwards compatible with SATA: SATA II and SATA 6 Gb/s devices can be connected to a SAS controller, but SAS devices cannot be connected to a SATA controller. The latest implementation of SAS provides data transfer at speeds up to 12 Gb / s per line. 24Gb/s SAS specification expected by 2017
SAS combines the advantages of SCSI interfaces (deep sorting of the command queue, good scalability, high noise immunity, long maximum cable lengths) and Serial ATA (thin, flexible, cheap cables, hot-pluggability, point-to-point topology, which allows achieving better performance in complex configurations) with new unique features such as advanced connectivity topology using hubs called SAS extenders ( SAS expanders), connecting two SAS channels to one (both to increase reliability and performance), work on one disk with both SAS and SATA interfaces.
In conjunction with new system addressing, this allows you to connect up to 128 devices per port and have up to 16256 devices on the controller, without the need for any manipulation of jumpers, etc. Removed the 2 Terabyte limit on LUN space.
The maximum cable length between two SAS devices is 10 m when using passive copper cables.
Actually, the SAS data transfer protocol means three protocols at once - SSP (Serial SCSI Protocol), which provides the transfer of SCSI commands, SMP (SCSI Management Protocol), which works with control SCSI commands and is responsible, for example, for interacting with SAS extenders, and STP (SATA Tunneled Protocol), which implements support for SATA devices.
Produced in this moment have internal connectors like SFF-8643 (it can also be called mini SAS HD), but you can still meet connectors like SFF-8087 (mini SAS), which has 4 SAS channels.
The external interface option uses the SFF-8644 connector, but the SFF-8088 connector may still be found. It also supports four SAS channels.
SAS controllers are fully compatible with SATA drives and SATA cages/backplanes– the connection is usually made with cables: . The cable looks like this:
SFF-8643 -> 4 x SAS/SATA
Usually SAS cages / backplanes have SATA connectors on the outside and you can always insert regular SATA drives into them, so they (such cages) are usually called SAS / SATA.
However, there are reverse versions of such a cable for connecting a backplane with internal SFF-8087 connectors to a SAS controller that has regular SATA connectors. These cables are not interchangeable with each other.
SAS drives cannot be connected to a SATA controller or installed in a SATA cage/backplane.
To connect SAS disks to a controller with internal SFF-8643 or SFF-8087 connectors without using SAS cages, you must use a cable like SFF-8643->SFF-8482 or SFF-8087->SFF-8482, respectively.
Existing versions of the SAS interface (1.0, 2.0, and 3.0) are compatible with each other, that is, a SAS2.0 drive can be connected to a SAS 3.0 controller and vice versa. In addition, the future version of 24 Gb / s will also be backwards compatible.
Types of SAS Connectors
| Image | code name | Also known as | External/ interior |
Number of contacts | Number of devices |