In This Article
What is Direct Access Storage Device (DASD)?
A Direct Access Storage Device refers to a secondary storage device where each record has a unique address and a discrete location. Technologically, this refers to the storage devices developed by IBM to use in mainframe and microcomputers.
In simple terms DASD means non-sequential access to data which is fast since searching in successive locations is not required.
- DASD allows faster access, reading and writing because there is no need to search for the desired location sequentially.
- DASDs can be categorized into three groups based on their characteristics such as optical storage devices, flash storage, and magnetic disks.
- The Direct Access Storage Devices are ideal solutions for small businesses because these are quite low cost.
- These storage devices are easy to configure and set up, offer high performance, but do not offer scalability and central management.
Understanding Direct Access Storage Device
A Direct Access Storage Device is a general term and it refers to those particular storage devices that support ‘direct access,’ and hence the name.
Typically, the Direct Access Storage Devices can read or write directly from or to a particular place.
The data stored in the Direct Access Storage Device can be directly accessed by the users in a quick time.
This is due to the fact that they do not have to do it by progressing through the data sequentially.
It is this particular aspect of the Direct Access Storage Devices that makes them entirely different from the tape data storage technology, which is typically restricted to sequential access.
Historically, the term Direct Access Storage Device was used in minicomputer and mainframe environments.
Therefore, it is safe to say that the term Direct Access Storage Device is a generic term that is used to refer to all the peripheral computer storage devices that have almost the similar type of data access characteristic.
The same characteristic is also applicable to the core and the magnetic tape without any implication to the time scale.
However, things would be more or less logarithmic in that case.
Typically, the time taken to access by the core memories of the systems is not dependent on the addressing distance between the accessions but it is usually in the order of 1 to 10 microseconds.
With reference to the access time, that of the magnetic tapes is usually very linear in nature on a single reel of tape, which, typically, may extend over an addressing distance of about 3 to 4 million words or up to 15 million bytes.
Usually it is found that for a 90 KB/s tape, the typical slope of the timeline is about 10 seconds per byte.
On the other hand, as in the case of the Direct Access Storage Device, it has linear portions of its characteristics.
These are typically divided by vertical jumps.
However, the mechanical construction of the Direct Access Storage Device typically determines the height as well as the number of these jumps.
In fact, it is based on the height and number of these breaks in their characteristic that the Direct Access Storage Devices are classified even further, along with the concept of the distance between the addresses redefined.
Ideally, these dedicated storage devices are attached to the computer or to a server directly through a cable.
This can be any type cable that follows any of the following major protocols for connections:
- SATA or Serial Advanced Technology Attachment
- eSATA or External Serial Advanced Technology Attachment
- ATS or App Transport Security
- IDE or Integrated Drive Electronics
- SCSI or Small Computer System Interface
- SAS or Serial Attached Small Computer System Interface and
- Fiber channel.
Depending on the type of protocol followed, the speed and performance may vary a bit.
Typically, depending on the characteristics, the Direct Access Storage Devices can be categorized into three specific groups such as:
- Flash Memory
- Magnetic disks and
- Optical storage devices.
Each of these categories is explained below for your better understanding.
Flash memory is a specific type of non-volatile memory.
It stores data in small units called blocks and these should be erased before writing any data on the flash memory chip or before programming it.
Data can be stored for a longer period of time in a flash memory irrespective of whether the flash-equipped system is powered off or on.
There is a cell containing the storage transistor that also has a floating gate and a control gate.
Electrical charges are stored in the floating gate which controls the flow of it as well.
This is typically insulated with a thin oxide layer or dielectric material from the other transistors.
The design of the flash memory has evolved from EPROM or Erasable Programmable Read Only Memory and EEPROM or Electrically Erasable Programmable Read-Only Memory.
A NOR flash typically rewrites data at the byte level and erases data at the block level.
In NAND flash memory, however, the data is written at multiple-byte page level.
A flash memory is used extensively in consumer devices, industrial applications and enterprise systems to store and transfer data which follows the process known as Fowler-Nordheim tunneling.
It removes the electrons from the floating gate.
The advantages of using the flash storage systems include:
- It is faster read and write speeds
- It is small in size
- It is less susceptible to damage
- It is cheaper to produce and
- Its less power consumption.
However, the downsides include:
- Its limited number of erase and write cycles before failing
- Its lack of write-protection mechanism in many
- It is easier to lose due to small size and
- Its limited storage space as compared to hard drives.
The magnetic disks can be either fixed-head magnetic disk storage or movable-head magnetic disk storage.
- Fixed-head Magnetic Disk Storage
The Fixed-head variant typically looks like a large Compact Disc or Digital Versatile Disc.
The surface of the disc is covered with a magnetic film. These discs are usually formatted on both sides and contain concentric circles known as tracks.
The data is recorded one after the other in a serial manner on every track. There is a fixed read/write head to locate the necessary data.
These types of storage solutions are pretty fast but are quite expensive compared to the reduced storage space that they offer.
- Movable-head Magnetic Disk Storage
These types of storage solutions typically come with one floating read/write head over every surface of the disk.
The disk can be either a single platter or a stack of several magnetic platters.
Every platter in this type of storage comes with two separate recording surfaces except the top and bottom platters and has a fixed number of concentric tracks to record the data.
This number may vary from one manufacturer to the other but actually there are in excess of a thousand of these tracks on any given high-capacity hard disk.
On the other hand, in a disk pack platter design, every track on each surface comes with a specific number.
Track zero indicates the outermost circle on each surface and the highest number is in the center.
There is an arm that moves the two read/write heads between every pair of the surfaces in unison.
One head is for the surface below it and one is for the top of it.
These are placed on the same track but on their particular surfaces which creates a virtual cylinder.
Optical Storage Devices:
These are the disks that come with a plastic coating on which you can store text, music, or video digitally on very small regions.
These microscopic regions vary in reflectivity and there is a laser that scans the surface to read it.
Therefore, these devices are also called laser disks.
Typically, the optical discs contain one spiraling track divided in sectors of uniform size. This runs from the center to the edge of the disk.
The design of the optical disks has more sectors and therefore can store much more data in comparison to a magnetic disk of equal size.
There are different types of optical disc systems available in the market such as:
- CD-ROM or Compact Disc Read Only Memory, which allows reading the data but not writing or saving anything on it
- CD-R WORM or Compact Disc Write Once Read Many, which allows saving data on the disk only one time but reading it many times and
- CD-RW or Compact Disk Read Write, which allows both writing and saving data on the disk over and over again.
The advantages of Compact Discs are these are:
- Small and easy to carry
- Cheaper to make
- Quite fast in performance than magnetic tapes and floppy disks and
- Compatible with most computers.
However, the downsides include lower storage space and accessing speed as compared to HDDs.
These are also very fragile and get scratched quite easily if not handled with care.
The DVD technology is much similar to the CD technology in design, size and shape.
However, it can store as much data as 13 CDs and come with a laser of shorter wavelength.
The spirals can be wound tighter and make smaller pits.
Architecture and Working Process
The architecture of the Direct Access Storage Devices is pretty simple.
The host computer can access the direct attached storage of it directly or access the data stored in an external Direct Access Storage Device connected through a network to the storage servers.
In comparison, the other storage architectures such as Storage Area Network or SAN and Network Attached Storage or NAS solutions come with very complex architecture.
However, these solutions will offer performance benefits that the Direct Access Storage Devices cannot offer.
The Direct access Storage Devices are quite low cost in design which is why these storage devices are ideal solutions for the small businesses which are not likely to have an exceptionally increasing storage requirement in the foreseeable future.
Almost every computer today comes with a direct storage attached to it. This may be in the form of an internal storage drive such as the conventional and slower Hard Disk Drives or HDDs or the faster Solid State Drives or SSDs.
These drives are usually connected to it via a Serial Advanced Technology Attachment or SATA interface.
Also, the servers today come equipped with internal storage drives and these too are connected through the SATA, SCSI, SAS or any other high speed interfaces. This offers a much better and higher storage performance.
However, the Direct Access Storage Devices do not need to be internally connected to the host computer system in order to work.
It can be an external drive enclosure containing several drives or a single external drive connected to it through a USB, SATA, eSATA, SCSI, or SAS interface.
Whatever it is, the defining feature of their working process is that they are all controlled by only one single computer to which they are connected.
This means that any other computer that may need to access the data cannot access the Direct Access Storage Device directly.
Instead, it has to first communicate with the host computer that the storage device is connected to and then get permission to access the data stored in it.
Direct Access Storage Device in Data Structure
Data can be accessed directly by the host computer from the Direct Access Storage Device easily and quickly due to the efficient storing ability of these devices.
The data structure in these devices is usually in chunks and each of these data chunks is stored in a discrete location from the other data chunks in it. All these locations have a unique address.
The specific architecture allows the mainframe to access the input-output devices via different channels.
These channels act basically as a kind of secondary mini processor. These channels programs read from, write to, and also control the specific device.
There is a Count Key Data or CKD which refers to the data record that is addressable and stored on a CTR disc.
In the earlier days, in the 1970s to be precise, Fixed Block Architecture or FBA was introduced by IBM.
The devices with such architecture typically did not use the conventional CHR addressing at the programming level. Instead, these devices referenced the blocks of fixed length by number.
This was pretty similar to the segments in the mini computers.
In this type of arrangement, the storage underlying was not known to the application programmer, which, however, stored the data in fixed physical blocks of varying lengths of 512, 1024, 2048, or 4096.
This FBA in particular offered a lot of simplicity to several applications but more importantly, in addition to that, it also offered an increased throughput with a dramatic boost in the speed of operations.
When it comes to accessing, it is the programming interface routines and macros that are called Direct Access Method or DAM. These had different names and functionalities.
For example, the DOS/360 and successors through z/VSE supports sets of data on a Direct Access Storage Device through these following access methods:
- Logical Input Output Control System or LIOCS and
- Physical Input Output Control System or PIOCS.
LIOCS access methods can be further subdivided into:
- Sequential Access Method or SAM
- Indexed Sequential Access Method or ISAM
- Virtual Storage Access Method or VSAM and
- Direct Access Method or DAM.
Physical IOCS, on the other hand, only have the Execute Channel Program or EXCP to follow.
And, the OS/360 and successors through z/OS supports sets of data on a Direct Access Storage Device through these following access methods:
- Basic Sequential Access Method or BSAM
- Virtual Storage Access Method or VSAM
- Execute Channel Program or EXCP
- Basic Indexed Sequential Access Method or BISAM
- Basic Partitioned Access Method or BPAM
- Basic Direct Access Method or BDAM
- Queued Sequential Access Method or QSAM
- Queued Indexed Sequential Access Method or QISAM and
- Execute Channel Program in Real Storage or EXCPVR.
However, in MVS, right from the beginning with OS/VS2 Release 2 till the end through z/OS as well as EXCPVR, all these access methods used the advantaged STARTIO macro.
In these times the data cells and drum both have however disappeared.
Therefore the Direct Access Storage Devices are now synonymous to the disk devices.
All modern DASDs that are used in mainframes use large disk arrays most commonly and use RAID schemes.
You will typically find only a very few that come with single disk-drives, if you at all find that is.
The architecture and data structure enables the computer to point to the specific location directly in which the data is available and can access it through different access methods such as direct, indexed, and sequential.
However, this does not mean that the speed of accessing will be equal and fast in all, even if the precise position of the data is known.
This is because the speed of accessing the data mainly depends on the ability of the Direct Access Storage Devices.
For example, if you consider a tape drive wherein the exact data location is known, the speed will not be fast because even then the device has to follow the sequential access method due to its inherent design.
This means that the tape drive will have to go through all those locations in it that are prior to the particular location that is required to access.
Furthermore, the tape drives themselves are pretty slow in operation as opposed to the direct access disks which can spin very quickly and move the read and write head over the disk to the precise sector and track just in a matter of a few seconds.
However, these devices become inaccessible, unlike the Network Attached Storage devices once the system goes offline which these are connected to.
Direct Access Storage Device in Operating System
Normally, depending on the type of the Direct Access Storage Device and the operating system, the accessing methods may vary.
For example, for accessing the data the operating systems use a 4-byte comparative TTR or Track and Record.
However, there are other operating systems some other access methods such as:
- 8-byte extent-bin-cylinder-track-record block address or MBBCCHHR
- Channel programs address Direct Access Storage Device with a 6-byte seek address or BBCCHH and
- A 5-byte record identifier or CCHHR.
Here the different letters indicate different aspects such as:
- M signifies the extent number in the allocation
- BB signifies the Bin from 2321 data cells
- CC signifies the cylinder
- HH signifies the Head or track and
- R signifies the record or block number.
However, the 2321 data cell was terminated in January 1975.
In that case, the device itself and the addressing format were represented as CHR or CTR for the cylinder track record.
This is because the bin number was 0, always.
However, it may not be the same always and every time.
This is because a single block may have a number of logical or user records which are also called spanned records or partial logical records in some schemes.
Here, according to IBM, logical records refer to the data records that the programmers usually work with and blocks or physical records refer to the format of the Direct Access Storage Device.
Depending on the track limit, the physical records can be of any size but the thing to note is that there are a few devices that may come with a track overflow feature.
This will allow the device to break a large block down into segments of the size of the track within the same cylinder.
And, as in the queued access methods, such as the Queued Sequential Access Method or QSAM, the logical records are either blocked or unblocked as and when these are read from or written to an external media.
In BSAM or the Basic Sequential Access Method and other basic access methods it is the responsibility of the user program to do such jobs to access.
Fixed Block Architecture or FBA is typically supported by DOS/VSE and VM/370 and but not by MVS or by the descendant operating systems in the OS/360 line.
And, on the other hand, the processors equipped with FICON channels can access the SCSI drives through the FCP or Fiber Channel Protocol.
However, z/VSE and z/VM completely support FCP but z/OS provides limited support only via IOSFBA.
Advantages and Disadvantages of Direct Access Storage Devices
There are lots of advantages offered by the Direct Access Storage Device and some of them are:
You can expect to have a higher performance from a Direct Access Storage Device in comparison to other traditional storage methods. The performance is high and fast because data is transferred directly between the Input/output devices and the memory which saves the CPU from such botherations. The CPU can use this time saved for performing other major operations that do not need system buses.
It will offer faster and direct access since it is connected directly to the computer. This means that there will be no issues related to network connectivity and congestion.
However, there may be such issues if your computer tries to access data that is stored on another Direct Access Storage Device that is linked to a storage server through a network.
Configuration and Setup
The Direct Access Storage Device is very easy to set up and configure. Your computer may even be equipped with such direct attached storage devices internally that are ready to use right away.
And, if you want to use an external Direct Access Storage Device it will usually be ‘plug and play’ type. This means that it can be used simply by plugging it into a port, such as a USB port, that supports it.
Low Cost Option
Since the Direct Access Storage Device consists of only the storage device itself along with the drive enclosure, it is very cost effective in comparison to the other traditional storage solutions. This is because the DASD will not need any additional software to download or hardware to use to run and manage it.
Some other major advantages offered by the Direct Access Storage Device are:
- Higher availability
- Higher access rate due to the absence of Storage Area Network or SAN
- Expansion of storage capacity
- Data security
- Fault tolerance and
- Ease of administration and management.
However, everything is not good with the Direct Access Storage Devices and they also come with a few drawbacks in them just like any other computer peripheral device. Few of the demerits of a Direct Access Storage Device are:
Typically, a Direct Access Storage Device is pretty hard to scale. This is because there are limited options for it in terms of the amount of internal drive bays, the capability of the external Direct Access Storage Device, and the availability of exterior ports.
Also, as for the internal Direct Access Storage Device, if it needs upgrading, it will be needed to shut down the host computer during the process.
Possible Poor Performance
A Direct Access Storage Device connected to a computer may perform poorly especially when you need to share data stored in them with other computers on the same network. It will be pretty slow in providing the data mainly because the level of performance depends partially on the available resources of the host computer.
Also, when you share data through a DASD it may even reduce the performance level of the host computer. However, the effect may be significantly low if the Direct Access Storage Device is connected to a powerful server that is dedicated to storage.
Lack of Central Management
Typically, there is no backup or central management that may ensure that the data that is stored in the Direct Access Storage Device is well protected.
This is because the process is very much complicated and also involves a lot of cost in comparison to creating backups or arranging redundancy on the networked storage devices.
The primary reason behind it is that the network attached storage devices usually come with their own management, backup, and RAID software.
However, this may not be a major issue where there is a limited number of computers using a Direct Access Storage Device. However, it will become quite a significant issue when the numbers of computers proliferate with the growth of the organization.
A few other significant disadvantages of Direct Access Storage Device include:
- Inability of diverse groups to access data
- High administrative overheads
- Permitting only one user at a time
- Limited sharing
- Unsuitable for IT managers
- Inappropriate capacity utilization and
- The CPU rendered inactive for a long time during transferring data in burst mode.
Still, with all these demerits, the Direct Access Storage Devices offer much more benefits which makes them quite suitable to use for modern computing storage needs.
What is Direct Access Storage Device Used for?
A Direct Access Storage Device is typically used in servers as well as in personal computers as an internal storage.
Apart from that, a DASD can also be used as an external storage device such as a Universal Serial Bus or USB or as any other external hard drives.
Typically, in the Small And Medium-sized Businesses or SMBs, a Direct Access Storage Device can also be used as a file server.
On the other hand, in large organizations such as in the data centers these devices are typically used as a private storage that is attached to a dedicated server.
In addition to that, the larger enterprises can also use the Direct Access Storage Device sometimes with different types of networked storage systems such as SAN or Storage Area Network and NAS or Network Attached Storage device.
Typically, a Direct Access Storage Device is used where huge amounts of storage capacity is required along with high performance.
It is one of the most practical, low-cost, and preferred storage choices for the businesses that need to use simpler storage systems and do not want to share the data across the entire organization.
Is DVD a Direct Access Storage Device?
No, it is not. Ideally, a DVD or a Digital Versatile Disc will allow direct access to all its parts. For example, when you watch a DVD movie, much unlike the VHS videotape movie, you may hop, skip and jump to any particular scene on the disc.
Therefore, you may quite naturally think that it is a DASD-type of file storage.
However, a DVD is typically not supported by LVM or Logical Volume Manager, which is a type of storage virtualization that offers more flexibility while managing disk storage space in comparison to traditional partitioning.
And according to the AIX documentation of IBM, any storage device that does not support an LVM can be anything but surely not a Direct Access Storage Device.
What Are Not Direct Access Storage Devices?
Well, there is an ongoing debate as to which devices are to be considered as a Direct Access Storage Device and which are not to be considered as DASDs.
According to the definition, since the Direct Access Storage Devices can be fixed or removable storage devices that usually have a rotating disk or even a solid state disk, you may think that every storage device falls under this category.
This confusion is due to the inclusions and exclusions of the devices in this specific category.
For example, few magnetic and optical disc drives fall into this category but there are a few specific types of both that do not.
However, as said earlier, based on the AIX documentation of IBM and on the fact that whether or not the devices are supported by the Logical Volume Manager or LVM, this is not always true.
Typically, these following types of devices are not supported by the LVM and therefore are not considered to be a Direct Access Storage Device:
- CD-ROM or Compact Disk Read Only Memory devices
- DVD or Digital Versatile Disc drives and
- WORM or Write Once, Read Many devices.
Still, after all the things said, it is good to take note of an important aspect at this point that the Direct Access Storage Device uses an addressing system that helps in identifying distinct positions on the storage media.
This allows direct, easier and quicker access to a specific data without going through the unwieldy overhead that is associated with sequential access.
Now you know what the Direct Access Storage Devices are and what does not fall into that category, courtesy this article.
This knowledge will surely help you in the future if you have to make a choice between different storage devices to use in your computer.