Arithmetic logic units (ALU) perform arithmetic and logic operations on binary data inputs. In some processors, the ALU is divided into two units: an arithmetic unit (AU) and a logic unit (LU). In processors with multiple arithmetic units, one AU may be used for fixed-point operations while another is used for floating-point operations. In some personal computers (PCs), floating-point operations are performed by a special floating-point AU that is located on a separate chip called a numeric coprocessor. Typically, arithmetic logic units have direct input and output access to the processor controller, main memory and input/output (I/O) devices. Inputs and outputs flow along an electronic path called a bus. Each input consists of a machine instruction word that contains an operation code, one or more operands, and sometimes a format code. The operation code determines the operations to perform and the operands to use. When combined with a format code, it also indicates whether the operation is fixed-point or floating-point. ALU outputs are placed in a storage register. Generally, arithmetic logic units include storage points for input operands, operands that are being added, the accumulated result, and shifted results. 

Arithmetic logic units vary in terms of number of bits, supply voltage, operating current, propagation delay, power dissipation, and operating temperature. The number of bits equals the width of the two input words on which the ALU performance arithmetic and logical operations. Common configurations include 2-bit, 4-bit, 8-bit, 16-bit, 32-bit and 64-bit ALUs. Supply voltages range from - 5 V to 5 V and include intermediate voltages such as - 4.5 V, - 3.3 V, - 3 V, 1.2 V, 1.5 V, 1.8 V, 2.5 V, 3 V, 3.3 V, and 3.6 V. The operating current is the minimum current needed for active operation. The propagation delay is the time interval between the application of an input signal and the occurrence of the corresponding output. Power dissipation, the total power consumption of the device, is generally expressed in watts (W) or milliwatts (mW). Operating temperature is a full-required range.

Selecting arithmetic logic units requires an analysis of logic families. Transistor-transistor logic (TTL) and related technologies such as Fairchild advanced Schottky TTL (FAST) use transistors as digital switches. By contrast, emitter coupled logic (ECL) uses transistors to steer current through gates that compute logical functions. Another logic family, complementary metal-oxide semiconductor (CMOS), uses a combination of P-type and N-type metal-oxide-semiconductor field effect transistors (MOSFETs) to implement logic gates and other digital circuits. Bipolar CMOS (BiCMOS) is a silicon-germanium technology that combines the high speed of bipolar TTL with the low power consumption of CMOS. Other logic families for arithmetic logic units include cross-bar switch technology (CBT), gallium arsenide (GaAs), integrated injection logic (I²L) and silicon on sapphire (SOS). Gunning with transceiver logic (GTL) and gunning with transceiver logic plus (GTLP) are also available.

Arithmetic logic units are available in a variety of integrated circuit (IC) package types and with different numbers of pins. Basic IC package types for ALUs include ball grid array (BGA), quad flat package (QFP), single in-line package (SIP), and dual in-line package (DIP). Many packaging variants are available. For example, BGA variants include plastic-ball grid array (PBGA) and tape-ball grid array (TBGA). QFP variants include low-profile quad flat package (LQFP) and thin quad flat package (TQFP). DIPs are available in either ceramic (CDIP) or plastic (PDIP). Other IC package types include small outline package (SOP), thin small outline package (TSOP), and shrink small outline package (SSOP).