Special offer. See more information about Outbyte and uninstall instructions. Please review EULA and Privacy policy.

iTechGuides is reader-supported. When you buy through links on our site, we may earn an affiliate commission. As an Amazon Associate I earn from qualifying purchases. Learn more

Nusku is an in-progress Linux continuous profiler that Ali Amer is building in Zig, with eBPF intended to observe what running programs are doing. Its goal is to make activity such as CPU use, memory allocation, waiting, system calls, and network retransmissions visible while a program runs. Those are project aims described by Amer, not verified features of a released tool.

What Nusku is intended to do

Amer describes Nusku as a way to inspect a running program rather than rely only on assumptions or investigate after a performance incident. The planned view reaches beyond CPU activity: it is meant to help identify where functions spend CPU time, where memory allocations originate, when threads wait on locks or I/O, how long system calls take, and where packet retransmissions contribute to network delays. These capabilities come from the project introduction; it does not provide independent testing or evidence that they are all implemented.

The intended workflows include observing a process already running on a Linux machine or launching a process through Nusku. Amer also describes local observation and streaming a live view from a remote machine over a secure connection. These should be read as planned uses, not confirmed release behavior. The project introduction does not establish Nusku’s release state, supported kernel versions, required privileges, setup steps, or runtime overhead.

What’s actually slowing this PC down?

Pick the symptom - the matching free tool is one click away.

Special offer. See more information about Outbyte and uninstall instructions. Please review EULA and Privacy policy.

How the proposed architecture is organized

The Agent comes first

The planned Agent would sit alongside a target process and collect observations through eBPF. Amer says development is currently focused on this component, describing the present scope as “the Agent and nothing else.” That makes Nusku a project still being built, rather than a complete profiling system with a documented end-to-end workflow.

A Controller is a later component

A planned Controller would list running targets, let a user select one, and manage profiling sessions. Amer describes it as useful when monitoring multiple processes or working remotely, but says it is not the current development focus. The project introduction therefore does not establish that target selection, session management, or remote operation is available today.

Why Amer chose Zig and eBPF

Amer’s stated reason for using Zig is an interest in systems programming and direct control over memory and allocation without a garbage collector. He presents eBPF as a way to observe activity at the kernel level without injecting code into the target application. These are the author’s design reasons; they do not demonstrate that eBPF is the only suitable approach to profiling or that Nusku’s implementation has already achieved its goals.

How Nusku fits alongside an existing Linux profiler

One concrete point of context is lightswitch. Its repository describes a Linux profiler for on-demand and continuous on-CPU profiling, with a live terminal flamegraph and saved profile formats. The repository identifies it as written in Rust with C for eBPF.

Special offer. See more information about Outbyte and uninstall instructions. Please review EULA and Privacy policy.

That is a reference point, not a head-to-head comparison: the available project descriptions do not establish equivalent tests or implementation status. Nusku’s stated scope is broader in the kinds of signals it aims to expose, while lightswitch documents CPU profiling behavior. Nusku’s maturity, deployment requirements, compatibility, and overhead cannot be compared reliably from the project introduction alone.

Independent reader supportYour contribution helps us test, update, and keep practical guides available for everyone.Support on Ko-Fi

What to watch as Nusku develops

  • Implemented behavior: whether the planned CPU, allocation, waiting, syscall, and network views become available in a usable Agent.
  • Operational details: documented kernel support, permissions, setup requirements, and measured overhead.
  • Workflow coverage: whether local process selection, launching a target, remote streaming, and the planned Controller are delivered.
  • Project direction: Amer presents learning and depth-first development as goals and says there is no current plan to monetize Nusku.

Amer’s project introduction appeared on DEV Community with a publication date shown as October 4, but no year visible in the article. It is best treated as a statement of intent and development direction, not a current compatibility or release guide.

Product prices and availability are accurate as of the date/time indicated and are subject to change. Any price and availability information displayed on Amazon at the time of purchase will apply.