Latency

Community Submission - Author: John Ma

Latency in computing refers to the time delay between an input and the received output. It is inherent at all levels of computing, from the IO latency between the user and the computer to the network latency as data and information travel from a computer to servers around the world. In cryptocurrencies, latency can refer to two different time delays. The first is the latency in the network of a blockchain, and the second is the latency on an exchange. 

Blockchain network latency is the time between submitting a transaction to a network and the first confirmation of acceptance by the network. After the first confirmation, the transaction becomes more final as more blocks are added beyond the initial confirmation. In a payments system that hopes to gain widespread adoption, low network latency is paramount. The time between payment at a cashier and the confirmation of the payment is a point of user friction if it grows too long. 

The latency of an exchange is a measure of their ability to process and execute large volumes of transactions in their order books. It is common for day traders to utilize bots to automate most of their trading volume, which means the bots are placing and canceling an incredibly high volume of orders every second. An exchange that has low latency and high throughput can process these orders in a timely fashion, and thus the day trader (via the bot) can take more considerable advantages of swings in asset prices. Conversely, an exchange that has high latency will process orders with a time lag behind the evolving asset price, which results in wrongly priced orders and opportunities missed for the day trader.