Glossary

Glossary of terms in computing

Automation : This refers to the set of tasks that have been specified in a way that they can run without manual intervention. This comes in very useful in test setup, execution, tear down, results store, and reports making.  Frequently such tasks are delegated to a test harness.

Balancing :  This refers to sharing the load between one or more processors or units of control such that there is a distribution of work resulting in more concurrency and faster execution as well as fault tolerance.

Cloud Computing : This refers to the use of one or more virtual machines in a hosted environment where the provider of VM and its network,  the blades and hardware are taking away the onus of maintaining the application on a proprietary system while assuring high availability and geographic reachability.

Degree of Parallelism: This refers to the number of concurrent unit of contol invoked in the execution of a task usually by a partitioning of data.

Execution  This refers to the actions needed to complete a task. The actions are usually outlined ahead together with the context and data with which to operate on and in a tightly controlled environment.

Flow diagrams : This refers to either the control or the data flows in a system where the system interaction and vulnerabilities can be studied by seeing how the data changes and with what controls as it goes through the system. This is one way to white box a system in that its internal workings are made clear with examples of data and control. Such diagrams are useful for threat analysis.

GigaByte: This is a unit of storage equal to 10 power 9 bytes and represents a way to describe the size of say files or databases.

Hashing : This is a technique to make lookups faster by keeping track of where insertions were made for entries in a collection so that they can be retrieved with a fixed cost. Typically, a range of hash values is used and a hash is computed on the object given and then put into the category.  The hashing is very effective when a good hash function is used otherwise chaining and linear search may result.

Instruction : This is a specific task specified to the processor and can be considered ‘bare metal’ or ‘native’ in that it is independent of any programming languages and very much tied to the hardware for which it is specified.

Jump : This is a very specific instruction given to the processor to move from the next instruction in a sequence to something outside the sequence.

Keys, public and private : This refers to cryptographic key pairs usually used for encryption such that only the sender and receiver can interpret the encrypted traffic. The private key is also called the secret and is used for decrypting while the public key can be shared.

Locking: This refers to the locking of a set of instructions or resources such that they provide mutual exclusion so as to prevent clobbering by interfering reads or writes. Interestingly, Lock free implies the guarantee of atomic reads and writes without conflicts by tightly controlling the sequence in which they appear but without the use of locks.

Monitoring: This is ensuring continuous availability of say a web service by frequently checking for some heartbeat events or directly calling the service periodically.

Nightly run: This refers to running large analytical workloads such as OLAP on system when there is less chance for conflicts with update statements.

Proxy: This is a software entity that sits between a sender and receiver and usually passes the traffic . Proxy can be either transparent or opaque and are commonly used with http and https.

Query: This is the language of retrieving data from a data store and can be both data definition as well as data manipulation in nature. They can be as granular as a batch or statement in T-SQL

Resource: A resource is known by its cost and is helpful with management.  It can be used to refer to memory or cpu required fo r execution and can loosely be used for humans assigned to different tasks.

Scale out/up:  A scale out is used to refer to additional servers added horizontally to say  a cluster  where as a scale up tries to increase the capacity of a single server in terms of memory or processors

Ticket: A ticket is a snapshot that captures enough information to identify an incident and for tracking. It can be associated with a user on whose behalf it is created and tracked as it is worked upon and closed.

User: A user to a computer system knows only the functionality and not the mechanics of the system. In other words, the user should not be expected to figure out whether a system failed to complete a specified action or what the remedial actions are.  The user only sees the system as a tool for solving a processing.

Violation, access:  Access violation is a bad address specified to the processor and most commonly is a null reference.

Zero: A zero is a state of a bit in which there is no signal and is a convenient test for the Boolean state false. It  provides a comparison against  a signal.