In-Memory Computing
- (The University of Chicago -Alvin Wei-Cheng Wong)
- Overview
In-memory computing (IMC) is a technique that runs computer calculations in computer memory, such as RAM. It's a type of middleware software that allows data to be stored in RAM and processed in parallel across a cluster of computers.
Here are some advantages of IMC:
- Speeds up data access: Data stored in RAM is available instantaneously, eliminating the need to retrieve it from disk and move it into RAM before processing.
- Scalability: IMC solutions can deliver real-time application performance and massive scalability.
- Energy efficiency: IMC uses the physical attributes and state dynamics of memory devices to perform computational tasks with very high areal and energy efficiency.
IMC is typically used for large-scale, complex calculations that require specialized systems software to run the calculations on computers working together in a cluster. Some examples of use cases that require high-speed processing include: Payment processing, Fraud detection, Predictive maintenance, Algorithmic trading, Self-driving cars.
Some disadvantages of IMC include:
- Volatility: Memory is volatile, which means if the system crashes or loses power, all data in memory can be lost.
- Cost: Memory is more expensive than disk storage.
[More to come ...]
