htop because it's much more user-friendly than top, has the feature of sending all kinds of signals to processes, has mouse support and it generally looks good. Not a fan of btop at all. Idk how to use it and I don't like the UI. I personally love the idea of no bloat. It's just such a nice little philosophy. Sometimes I even want to use a CLI only computer tbh. Though htop weights only a few kilobytes and it has features top doesn't have so I don't consider it bloat. I had it on my server as well
post
Uh, temperatures, that's nice.
I'd really like one of these to include GPU stats (I know, there's nvtop or whatever it's called), GUI apps can do it (Mission Center and a KDE system monitor widget), but I've not seen a CLI program include that ...
btop has GPU stats in recent versions.
Bottom gang unite
It's no burden. Don't overthink it. Use whatever you like.
htop is my go-to these days. It tells me what I need to know, and it's just nice to look at.
I've given both htop and btop a spin, and I have to say that I really prefer htop. It offers a prettier interface and more features than top, while still feeling less bloated than btop to me. So yeah, it's definitely my go-to choice!
btop for bling
htop for practical utility
top for minimalism, availability, reliability
htop on our vms and clusters, because it's in all the repos, it's fast, it's configurable by a deployable config file, it's very clearly laid out and it does everything I need. I definitely would not call it bloated in any way.
My config includes network and i/o traffic stats, and details cpu load type - this in particular makes iowait very easy to spot when finding out why something's racking up big sysloads. Plus, it looks very impressive on a machine with 80 cores...
My brain can't parse top's output very well for anything other than looking for the highest cpu process.
But - ymmv. Everyone has a preference and we have lots of choice, it doesn't make one thing better or worse than another.
htop gives me enough info without being too busy or slow, it's also in basically every OS repo by default so no complicated install.
The other ones can look awesome, but they're often harder to get info from quickly due to being too cluttered.
I like htop because it has nice CPU graphs and a good tui for navigating. Top is a bit too obtuse for a new user, especially since CPU time is measured per core and not per the entire CPU. Plus I never figured out how turbo boost plays a roll in those percents.
I haven’t gotten around to messing with btop, but it seems like more of what I like.
Also fuck the “muh bloat” people. I have an i9 and 32 gigs of ram. I don’t care that a monitor util takes 1/10th of a second longer to launch and uses 1MB more of ram.
Also fuck the “muh bloat” people. I have an i9 and 32 gigs of ram. I don’t care that a monitor util takes 1/10th of a second longer to launch and uses 1MB more of ram.
Maybe you only use those tools on your desktop but on a cloud server with only 1-2GB of RAM you really don't want your monitoring to take up some significant percentage of that. Especially when you are debugging things like OOM conditions already.
If I was that memory- and cpu-constrained I would be using other tools such as memstat, iostat, and cpustat.
-
Why are you using top instead of ps if you're worried about memory?
-
Containerise it, and you can debug locally
-
It's not what the person you're replying to is talking about. You're using a slightly better tool for a specific job, they're talking about people who won't use htop/btop on their own machine because BLOAT.
you really don’t want your monitoring to take up some significant percentage of that
except it doesn't - both htop and btop use <30 MB
and if 20MB makes a difference, you don't need a different top, you need a different machine
"bloat bad" people are just obnoxious
btop because pretty colors :3
i still need to learn how to use top well though, just in case that's the only option some day. if all else fails i just resort back to ps and (p)kill.
htop because pretty colors and graphs.
Or top because it's like muscle memory now.
I like htop more, but btop has better colors and graphs (imo)
I tend to go with htop purely out of habit. btop is better but I simply don't think to use it.
Why do you think that? After this post, I will try out both of them but maybe eventually I will still just use top out of, same as you bro, habit.
I'm more of a bottom, if you know what I mean.
Htop, but only because its what I've always used and have no need to change at the moment.
I'm more of a bottom guy myself
btop for system resource monitoring, htop for actually finding and killing processes
Htop is completly customizable for how the sections of data are displayed. it is a bit convoluted the first time you start, but then it makes
...sense.
Much better to quickly determine CPU and memory load.
atop, especially because you can take snapshots over time of what the system was doing and use it to backtrack when bad things happen.
I like btop because of its ease of use and modern gui. When I open top or even htop it feels like I'm using something designed for a dumb terminal from the 70s. When opening btop it feels like something designed for how computers are used now and not 50 years ago.
Also to my knowledge It's the only full system monitor to include GPU monitoring (while other GPU monitors exist they usually only monitor the GPU and not the whole system)
bpytop, it's just too pretty to pass up.
I thought btop replaced it. Didn't know that was still around
Edit: looking at the github: it isn't around anymore 😅
Btop, it's pretty. Htop when I'm lazy or working on a system that's bare bones.
top
Because it exists in nearly every environment I might need to check usage. From my desktop, through laptops, lab machines, routers, embedded systems, IoT to cloud, I don't have to keep the muscle memory of more than one app.
They’re different tools with different purposes. What you’re asking is like “which do you prefer, hand driver, box/open end wrench, socket wrench or impact driver?”
Ps and top can be used to very easily figure out and address when processes are screwing up. Atop, htop and btop can be used to directly view stuff hardware reports in real-ish time so you can figure out if a process has stopped being “stepped” across cores, a disk has stopped responding in time or when there’s a lot of network traffic.
As utilities they operate within fundamentally different scopes, to the point with btop of being extremely zoomed out macro pictures that are helpful when taking in abstract information about a system.
top 50 comments