Archives: Wiki

broadband

Broadband in telecommunications is a term used to refer to any technology that utilizes a “broad” band of frequencies to transmit data.

When used in the context of internet services, broadband refers to practically any technology newer than dial-up internet. These include technologies such as DSL (Digital Subscriber Line), Cable, Fiber Optics, Wireless (Mobile Broadband), Satellite, or BPL (Broadband over PowerLines). Currently, there are no official specifications that define broadband.

coil whine

Coil whine is a phenomenon in electronics where electromagnetic coils within an electrical circuit exhibit a high pitched sound as electricity passes through the circuit. Typically, this is most often heard in high performance computer components such as graphics cards and processors, but is actually very common in other electronics such as fluorescent light ballasts.

While annoying and sometimes a bit scary, coil whine is simply a result of sounds made as electricity passes through electrical coils. This is typically a result of poor engineering as these sounds can be minimized by utilizing coil dampening techniques. Rest assured however, no damage will occur to your system.

CPU

The CPU, also known as the Central Processing Unit, is one of the core components in a computer that includes all the logic needed to process programming instructions. CPUs are often also referred to as processor or microprocessor. Some people erroneously refer to the entire computer as the CPU.

CSV

CSV, or Comma-Separated Values, is tabular data expressed in plain text separated by commas, and is most commonly found in a CSV file. CSV files are typically created by a computer program or database in order to export its data to another computer program or database.

The reason why CSV files are so common is because CSV files are simply plain text and can easily be read by any application such as Microsoft Excel.

 

CSV Example

Tabular Data

CompetitorsPlan 1Plan 2Plan 3
Competitor A$10$20$30
Competitor B$15$25$35
Competitor C$20$30$40

CSV Data

Competitors, Plan 1, Plan 2, Plan 3
Competitor A, $10, $20, $30
Competitor B, $15, $25, $35
Competitor C, $20, $30, $40

DHCP

DHCP, or Dynamic Host Configuration Protocol, is a technology designed to allow clients to request a lease of an IP address from a pool (also know as a table) of available IPs at the time of the request. DHCP can also supply other information such as the subnet mask and default gateway.

The primary use of DHCP is to simplify the work of the network admin. It offloads the work of having to manually assign IPs and prevents IP conflicts among clients. It is an essential feature in both small home networks where consumers are unfamiliar with complicated network setup and large networks that have hundreds and thousands of clients.

FC SAN

FC SAN, or Fiber Channel Storage Area Network, is a type of SAN which uses fiber channel to connect servers and the SAN. Fiber channel is an ultra high speed technology designed for transmitting data at data up to 128Gb/s.

hysteresis

Hysteresis in mechanical keyswitches occurs when the reset point on the switch is higher than the actuation point. For example, the Cherry MX Blue switch actuates at around 2.5mm from the top, but its reset point is only 1.5mm from the top. This means that the user would have to release the key beyond the actuation point before another keypress can be registered. Hysteresis is one of the main reason why many gamers prefer linear switches since the reset point is almost exactly where the actuation point is.

IOPS

IOPS, or Input/Output Operations per Second, is a storage performance metric which measures how many reads and writes are occurring every second. As smaller files take less time to transfer compared to larger files, the IOPS metric is typically used to measure transfers of 4K or 8K file sizes. Larger filesize reads and writes such as 128K, 512K and larger are typically measured by throughput such as MB/s (Megabytes Per Second) or GB/s (Gigabytes Per Second).

M.2

Plextor M6e M.2 SSD

M.2, which was originally known as NGFF (Next Generation Form Factor), is a card-type form factor developed by the standards groups SATA-IO and PCI-SIG. The specification was designed to replace the aging mSATA form factor which was limited by its physical size inflexibility as well as its supported interfaces.

Form Factor

M.2 is currently most commonly found in SSDs, or Solid State Drives, in the M.2 2280 form factor, which is 22mm wide and 80mm wide. SSDs can also be commonly found between 30mm to 110mm in length however, these are typically OEM solutions for the shorter drives and enterprise drives for the longer ones. However, the specification is very flexible in terms of physical sizing with drives that could be as small as 1630 (16mm wide, 30mm long), as large as 221110 (22mm wide, 110mm long) or as wide as 30mm in the case of the M.2 3030.

Interfaces

The biggest benefit of the M.2 form factor however, is its ability to be wired for many different interfaces. This includes SATA, USB, and PCIe. PCIe is expected to be the successor to the SATA interface as it’s highly scalable and is a mature technology already used by expansion cards such as graphics cards and storage adapters. PCIe also has the benefit of communicating using different storage protocols, allowing it to utilize the newer NVMe (Non-Volatile Memory Express) protocol which is expected to succeed AHCI (Advance Host Controller Interface) in SSDs.

machine learning

Machine learning is the science of teaching machines to make decisions on their own based on data presented without being explicitly programmed to do so. Typically, this is done through studying patterns in the data.

An example of machine learning is image recognition. For example, non-machine learning code may define a stop sign as an object that has an eight-side shape, is red, and has the word STOP on it. With machine learning, the system is simply shown lots of images of stop signs until it’s able to recognize any stop sign.

Within the machine learning field, two of the most basic types of algorithms are supervised learning and unsupervised learning.

Supervised Learning – In supervised learning, the algorithms are designed so that systems are essentially taught what to do in certain situations. After learning what to do, the system will then make decisions based on what it’s learned.

Unsupervised Learning – In unsupervised learning, the algorithms are designed so that the system is designed to take in data, interpret the data on its own and make decisions based on its own interpretations.