Dynamic random access memory (DRAM) is a type of semiconductor memory that is typically used for the data or program code needed by a computer processor to function. DRAM is a common type of random access memory (RAM) that is used in personal computers (PCs), workstations and servers. Random access allows the PC processor to access any part of the memory directly rather than having to proceed sequentially from a starting place. RAM is located close to a computer's processor and enables faster access to data than storage media such as hard disk drives and solid-state drives.
How does DRAM work?
Memory is made of bits of data or program code that are arranged in a two-dimensional grid. DRAM will store bits of data in what's called a storage, or memory cell, consisting of a capacitor and a transistor. The storage cells are typically organized in a rectangular configuration. When a charge is sent through a column, the transistor at the column is activated. A DRAM storage cell is dynamic, meaning that it needs to be refreshed or given a new electronic charge every few milliseconds to compensate for charge leaks from the capacitor.
The memory cells will work with other circuits that can be used to identify rows and columns, track the refresh process, instruct a cell whether or not to accept a charge and read or restore data from a cell.
DRAM is one option of semiconductor memory that a system designer can use when building a computer. Alternative memory choices include static RAM (SRAM), electrically erasable programmable read-only memory (EEPROM), NOR flash and NAND flash. Many systems use more than one type of memory.
Types of DRAM
There are many types of DRAM that can be used in a device. Some examples include the following:
- Synchronous DRAM (SDRAM) syncs memory speeds with CPU clock speeds, letting the memory controller know the CPU clock cycle. This allows the CPU to perform more instructions at a time.
- Rambus DRAM (RDRAM) was more widely used in the early 2000s for graphics cards.
- Double Data Rate SDRAM (DDR SDRAM) almost doubles the bandwidth in data rate of SDRAM by using double pinning. This process allows for data to transfer on rising and falling edges of a clock signal. It has been available in different iterations over time, including DDR2 SDRAM, DDR3 SDRAM and DDR4 SDRAM.
- Fast page mode DRAM (FPM DRAM) gives higher performance than other DRAM types through focusing on fast page access.
- Extended data out DRAM (EDO DRAM) improves the time to read from memory on microprocessors, such as the Intel Pentium.
Major DRAM manufacturers include Samsung, Rambus, PNY Technologies and SK Hynix.
Types of DRAM packages
There are two main types of DRAM packaging: single inline memory module (SIMM) and dual inline memory module (DIMM). Single inline memory module packaging is considered obsolete now and was used in the 1980s to 1990s. SIMMs came in 30 and 72 pin sets and typically had 32 bit data transfer rates. DIMMs, on the other hand, are commonly used now and are dual inline -- meaning that they have pins on both sides of the chip. DIMMS commonly have 168 pin connectors -- or more -- and support a 64 bit data transfer rate.
DRAM package types for DIMMs are set as different integrated circuit architectures. Some of these include the following:
- Unbuffered DIMMs (UDIMMs) are commonly used on desktops and laptops. These cost less and run faster, but are less stable.
- Registered DIMMs (RDIMMs) are commonly used with servers. These are more stable and reduce strain on a CPUs memory controller.
- Fully buffered DIMMs (FB-DIMMs) are used in larger memory systems. These are more reliable since they can improve error detection methods and maintain signal integrity.
The main advantages of DRAM include the following:
- Its design is simple, only requiring one transistor.
- The cost is low in comparison to alternative types of memory such as SRAM.
- It provides higher density levels.
- More data can be stored using DRAM.
- Memory can be refreshed and deleted while a program is running.
The main disadvantages of DRAM include the following:
- Memory is volatile.
- Power consumption is high relative to other options.
- Manufacturing is complex.
- Data in storage cells needs to be refreshed.
- It is slower than SRAM.
DRAM vs. SRAM
DRAM is a successor to SRAM. Memory designers reduced the number of elements per bit and eliminated differential bit lines to save chip area in order to create DRAM. As a result, DRAM is less expensive to produce than SRAM.
But SRAM retains some advantages over DRAM. SRAM does not need to be refreshed because it operates on the principle of switching the current flow in one of two directions rather than holding a charge in place within a storage cell. SRAM is generally used for cache memory, which can be accessed more quickly than DRAM.
SRAM is capable of byte-level reads/writes, and is faster at reads/writes than DRAM. DRAM writes data at the byte-level and reads at the multiple-byte page level.
Power differences vary based on whether the system is in active or sleep mode. DRAM requires less power than SRAM in an active state, but SRAM consumes considerably less power than DRAM does while in sleep mode.
One of the first uses of DRAM was in a Toshiba calculator in 1965 -- using a capacitive form of DRAM that was made from bipolar memory cells. That same year, IBM created a 16 bit silicon memory chip. However, at this time, the bipolar DRAM that was in use could not compete against magnetic-core memory. This stayed true of DRAM until the invention if the metal-oxide-semiconductor field-effect transistor (MOSFET), which lead to the metal-oxide-semiconductor DRAM -- or MOS DRAM. The patent for MOS DRAM was granted in 1968. 1969 saw Intel develop DRAM that used a three transistor cell.
Intel improved on their DRAM product with the Intel 1103 in 1970, seeing commercial use. Around this time was when MOS memory began to make more of a market presence compared to magnetic-core memory.
1973 also saw the invention of the Mostek MK4096, a 4 Kb DRAM. This was the first DRAM that incorporated multiplexed row and column address lines. The Mostek MK4096 could fit into small packages with a small pin count, since it halved the number of address lines required.
In 1992, Samsung developed SDRAM, which had a capacity of 16 Mb.