Thursday, November 21, 2024
Laptops

4 Best Laptops For Virtualization 2024 (10-50 VMs ATSM)

A laptop optimized for virtualization is generally needed in either of these four instances: 

  1. To run software that’s limited to one Operating System
  2. Developing and testing software for different OSs: Linux , Windows, OSX, Chrome OS, etc.
  3. Penn-testing.
  4. Labs for IT.

For the first instance..

You only need one virtual machine. Generally, a linux Distro on a Windows hypervisor (fancy word for VM software). Any laptop with 8GB RAM should be enough for this.

Thus you don’t need to do much research as most machines will do.

Even a 300 dollar  laptop  can run ONE VM with no hiccups

For the second instance..

You’ll probably need to run 3-4 VMs (they may or may not all be Windows based virtual machines) for which the RAM requirement goes up to 16GB or more.

For the third and fourth instance…

You will likely need to run anywhere from 2 to 50 VMs .

When you get started, Penn-testing will usually require you to run 2 as well (Kali Linux +VM) when getting started.

Running Kali Linux VM requires less RAM than a Windows VM. The latter may require more than 12GB RAM especially for software testing.

But as you dwelve deep into the IT field you may need to run up to 20 VMs. This is approx how much you’ll need for a home lab or (lab for practice) in the IT field. Either way, this means having at least 32-64GB RAM and of course the CPU with the largest number of cores you can get your hands on.

As for myself…

I’ve only used VMs for code testing (linux packages) in 2 VMs simultanously but I’ve built a PC for a friend who needed to run 20 VMs (IT purposes)  while trying to keep it on a budget so I have a pretty good idea of how much RAM, CPU resources and disk space VMs can gobble up. 

[ez-toc]

Best Laptop Specs for Virtualization

Based on that, here’s my recommendations when it comes to hardware for virtualization. 

I’ll make it very brief here but you can head over to the last section for all the details (and benchmarks). Also if you’re not adept with computer jargon, check my posts on the side bar. 

Note that you can just skip this section and head over the best laptop for virtualizations if you want!

RAM

As of 2024,  most CPUs have at least 4 cores (as we’ll talk about next it is enough for 7 VMs simultaneously) so RAM will be the main bottleneck.

8GB: This may seem low but it does the job for a couple of VMs ~4 . I’ve used this much for software testing (you don’t want to turn on and off a VM but rather have most OS running in the background).
16GB-32GB: minimum for 10 VMs (~10) and if you’re going to get into Virtual Labs/ Penn Testing simulations. You’ll start with 2 VMs (Linux + Win) but will later on run a dozen (thus 16-32GB is future proof).
+64GB: Only useful for very large lab set ups and found on SOME gaming laptops and workstation laptops.

CPU

Low #of VMs: Any modern CPU will be fine (2-3VMs).
High # of VMs: Pick the CPU with the highest # of threads you can afford. After that, clock speed also allows you to to run more VMs.

As a rule, you run x2 VMs / Thread efficiently. If you want to run more VMs/Thread, more clock speed helps

Note: Doesn’t matter what brand you get, both AMD & Intel support virtualization (Intel VT or AMD V). 

Storage

If you are bottlenecked due to insufficient RAM because you’re running dozens of VMs, the computer will use the  storage as  ‘extra RAM’. SSDs are many times faster than traditional HDDs, thus it is a MUST whether you’re running extra VMs or not.

All modern laptops have at least one installed. If adding a second storage, make sure it’s an SSD too. 

GPU

Graphics card do not increase VM performance nor help you run more VMs.

However, if running Penn Testin Labs (with several VMs). You probably will want to make good use of the GPU cores found on dGPUs. For example, to simulate attacks. If learning ethical hacking, you’ll also found the GPU cores useful (I discussed it in detail in my “best laptop for hacking” post in more detail. If you’re not doing any of these, you don’t need dedicated graphics.


TL;DR: For large # of VMs.. a 8-16 core CPU with high RAM support is required. Most laptops (modern) are upgradeable to 32GB. Bonus: DDR5 makes each VM faster

This excluded laptops below 350 dollars which either have a Core i3 or Ryzen 3 limited to 4-6 cores. Laptops below 300 dollars are good for 1VM however, as long as they have 8GB RAM. 

For large # of VMs: tiny laptops, 2-1 in laptops, touchscreen laptops are not recommended as they have RAM capped at 4-8GB that cannot be upgraded and low performance CPU for temperature control purposes.

Top 4 Best Laptops for Virtualization

Laptops aren’t just about hardware, since these laptops are going to be put through high loads and be your main tool of work. It’s reasonable to require:

  • Excellent build quality
  • Excellent Customer support
  • Great cooling systems (running dozens of VMs can raise temperatures quite a lot)
  • Easy Upgradeability (only a few are easy to upgrade and only a few support 48- 64GB).

Now before we get to the best laptops for virtualization you should know that:

  • Lenovo laptops usually have all the quality we look for at a reasonable price (they exclude the GPU which isn’t necesary in most cases).
  • Lenovo laptops (especially thinkpads) seem to be really designed for virtualization purposes, not only are they durable but they are very easy to RAM & Storage upgrade.
    • CPUs usually contain many threads while ditching useless hardware like fancy displays & GPUs.

Let us start with the most popular and ideal laptop for virtualization (~30) then we’ll go over cheaper options from other brands and lastly an option for one or two VMs (for running one  linux distro perhaps).


1. Lenovo ThinkPad T14 Gen 3 

Best Lenovo Laptop For Virtualization

 

   Intel Core i7 13th Gen i7-1360P 4P Cores + 8E cores (12 Cores/ 16 Threads) 3.7-5GHz.

  32GB RAM DDR5 (Soldered) Up to 64GB.

  NVIDIA® RTX™ A500 (4GB GDDR6 VRAM)

  1TB PCIe NVMe (Up to 2TB M2. SSD)

  16” FHD IPS  300 nits

  3.76 lbs

  10 hours

    No DVD Drive

  Windows 11 Pro

  WiFi 6

  Backlit Keyboard & Fingerprint reader

   x2 3.2 USB / x2 Thunderbolt 4 / x1 USB 4.0, 1x HDMI, 1x RJ-45, 1x SIM Card Slot

If you research long and hard for several days, you’ll come to the realization that the best laptops for virtualization are either going to be:

  • A) Gaming laptops (Because of high RAM & late gen CPUs)
  • B) Laptops tailored for the IT field which also means laptops with lots of RAM.

The problem with A (gaming laptops) is that they unnecessarily add a graphics card which increases the price significantly (though not always, #3 is an exception) and most people won’t need it as we talked about in the preceeding section.

The problem with B (business laptops) is that although, they have a pretty beefy and late CPU, enough for most VM users running a dozen, they lack the MOST powerful CPUs (Ryzen 9 & Core i9). 

I am assuming most people reading this post fall in the middle of running 2-3VMs to 50VMs. 

The best option of group B, is ALWAYS going to be (so far), a Lenovo ThinkPad due to high performance/money ratio. 

  Hardware

Now here’s the kicker, as of Late Dec 2023, Nearing 2024, there’s a Lenovo ThinkPad model which not only has the high RAM support and a 13th gen CPU but also a graphics card at pretty much no additional cost.

Let’s talk about each component individually.

13th Intel i7  1360P:

CPU P-cores E-cores Cores Threads Virtual Machines (Max) E Speed P Speed
i7 1360P 4 6 10 20 40 3.7GHz 5GHz

This is the latest generation of Intel CPUs, there’s no rumors of a 14th generation yet, this one just came out this year on laptops.

Now it may be a little confusing to grasp what the E cores and P cores mean but just focus on the #Threads part.

  • Although this may be a dedacore CPU, it only really has 16 threads which means 16 VMs should run at maximum efficiency.
  • That’s because the E cores only give 8 threads (there’s no hyperthreading) and the 4 remaining P cores give 8 (x2 each, there’s hyper threading) , totalling 16 threads.

This should let you run about 32 VMs maximum and with the high ram capacity (48GB-64GB depending on the model), you can assign 2GB RAM to each.

You can browse for other thinkpads with different stats at your discretion(portables, cheaper, more resistant – T series, etc). 

  • Clock Speeds on the 13th generation CPUs are only slightly faster than clock speed performance on 12th generation CPUs. The main difference is the addition of more E cores which take away a lot of background processing load from the P cores  so they can focus on high performance task.
  • Either way, your Hypervisor or your VM software, will automatically and efficiently distribute VMs depending on the number and loads you’re running on each.
  • If that isn’t the case, you can manually assign each VM to run on either the P core or E core, depending on how much performance you need.
  • Note that you may run into the issue that the VM software may only start using E-cores once you assign a certain # of threads/cores to the single VM. The solution is here. Basically just make sure windows is running on performance mode.

RAM set up : 16GB Soldered + 16GB added

RAM Slot 1 Slot 2 Total Max Windows VMs Linux VMs(1GB)
DDR5 16GB 16GB 32GB 48GB ~11(7) ~40

This laptop supports up to 48GB RAM. The 32GB RAM available is DDR5 thus according to Lenovo’s website as shown here, 32GB is soldered with an additional slot that supports 16GB RAM DDR5.

Soldered basically means that 16GB stick CANNOT be removed and you have to add an additional RAM stick to increase RAM, There are variations of this very same model that support 64GB but you have to buy them from the official website and the prices may increase significantly.

If you’re running dozens of VMs for simulation purposes, I suggest you try it with 32GB RAM, see how it goes, then add 16GB RAM. I think 48GB should be PLENTY for 98% of VM users, with the rare exception of IT specialization fields & penn testing, you may need 64GB or even use a desktop!

Also do note that the RAM here is DDR5, the fastest & the latest,  this translate to SIGNIFICANT more efficient Virtual machines as the data feeding process to the CPU is significantly faster.

GPU: RTX A500

VMs make no use whatsoever of dedicated graphics UNLESS you plan on launching software that requires a lot of parallel processing or run software with 3D graphics (perhaps software testing on a VM), if that’s the case, then yes it does become useful.

Now do not be mistaken, although this is a workstation GPU, its only as equally powerful as a 3050TI, that means it’s limited to small 3D models & gaming at 1080p. Though for software testing purposes and parallel processing it will work just as good as the highier tier GPUs.

Other ThinkPads:

If you look around Lenovo’s site or Amazon for other models, you may find cheaper models that have a 12th generation CPU or even a Ryzen CPU, while most of these laptops support at least 32GB. Only a few models support 48GB and even less models support 64GB. 

Which thinkpad to choose?

Whatever you choose really comes down to how many VMs you’re running and what exactly are you running on each.

In one of the most extreme scenarios, you may be running heavy loads on each VM for penn testing (attack simulations) or  in the case of IT purposes, for example, integration of Cisco Firewalls, switches with ESXi hosts require dozens of VMs. 

It’s really up to your needs. If you need a 64GB RAM GPU, please check out the following laptop below or laptop #3 on the list, which is a gaming laptop with 64GB RAM.

A last option would be a workstation GPU but they’re just so expensive due to the fact that they have a very beefy workstation GPU which is useless virtually for everyone running VMs, even in the extreme cases mentioned.

Lenovo Thinkpad P16S
PROS CONS
  • Latest 13th gen Intel i7 CPU
  • 16 Threads
  • Up to 64GB (other models)
  • Relatively portable & Lightweight
  • Ideal for dozens of VMs
  • Modern workstation GPU
  • GPU may be unnecessary
  • Low battery performance
  • 8GB is Soldered thus RAM limited to 64GB (this model) 

2. ASUS ROG Strix G16

Best Gaming Laptop For Virtualization

  Intel Intel Core i7-13650HX (6 E-cores , 8-P cores) 20 Threads 3.6GHz-4.9GHz

  64GB DDR5 RAM

  NVIDIA GeForce RTX 4060 8GB 140W MUX

  2TB PCIe NVMe

  165Hz 16” FHD

  5.51 lbs

  1.5 hours under high loads, 4 hours otherwise

  NO

  1 x USB Type-C (10Gbps, DisplayPort 1.4 + 140W Power Delivery), 1 x Audio Combo Jack, 1 x USB Type-A 3.2 Gen1 (5Gbps), 1 x E-Shutter Button, 1 x DC-in, 1 x RJ45, 2 x USB Type-A 3.2 Gen2 (10Gbps), 1 x HDMI 2.1; Wi-Fi 6 and Bluetooth 5.2

If you want high performance CPUs along with lots of RAM, your next best option is a gaming laptop. Although for the maximum hardware performance for an extremely large number of VMs would obviously be a desktop (more on this on the last section).

Anyways, let’s check out what a high-end gaming laptop has to offer for virtualization.

CPU: Core i7 13700H & Other high end CPUs.

CPU P-cores E-cores Cores Threads Virtual Machines (Max) E Speed P Speed
i7 13700H 6 8 14 20 40 3.6GHz 4.9GHz

By high end CPUs I mean Ryzen 9, Core i9, Ryzen 7 and workstation CPUs like the infamous Xeon CPUs. All these CPUs have the highest clock speeds and the highest (most efficient) number of cores or threads, the last part is crucial
and by that I mean a CPU like the Core i7 13700H has the first core running at 5GHz with turbo-boost and the remaining P cores (4.7) , which is pretty close to the advertised single core clock speed. 

The efficient cores run at much lower clock speeds nonetheless still fast enough to make significantly useful & efficient VMs that can run regular office software on each. 

This also applies to the Ryzen 9 and Core i9s, they have clock speeds evenly distributed among each core, pretty close to the advertised clock speed. Unlike the Core i7 137000H, these will have significantly higher clock speeds which is more evenly distributed among all cores (better multicore performance – basically clock speeds between cores are pretty close to each other).

What about Ryzen CPUs? Better for Virtualization?

The Ryzen 7 and Ryzen 9 or any Ryzen CPU of the latest generation will have clock speeds on each core running with EQUAL speeds. This means a very evenly distributed CPU performance on each VM, this CAN be useful for testing different versions of software on each VM to see how it runs under the same clock speed.

Overall, for all purposes, a similar equally faster (to the Intel Core ) Ryzen CPU will work much better to virtualization due to better multi-core performance (more resources for each VM) and the fact that hyperthreading FULLY works (with no bugs) on Ryzen CPUs. There’s also usually MORE cores (at least on desktops) available on Ryzen CPUs and on laptops the same # of Cores is found at a cheaper price.

So if you can find a Ryzen 7 CPU with the same stats as above at a cheaper price, GO FOR IT, UNLESS you need the high clock speed of 13th gen intel CPUs. 

RAM: 64GB DDR5 RAM

RAM Slot 1 Slot 2 Total Max Windows VMs Linux VMs (1GB)
DDR5 32GB 32GB 64GB 64GB ~15 ~60

Having the latest (7th) Ryzen 7 or Core i7 (13th) also means support for the latest and fastest RAM available. The fastest the Ryzen 7 supports runs at 6400MT/s and the Core i7 here supports up to 5200MHz. 

With 64GB RAM and a Core i7 , you could in theory run about x4 VMs per thread (16*2=32 VMs) and distribute 2GB RAM to each VM, though not enough for Windows 10 or 11, you could run 32VMs of Linux Distros. For the same number of Window VMs, you’d need a 128GB RAM (4GB is minimum for Windows). You can find those laptops on very expensive workstation laptops approaching 3k or even 4k dollars!

GPU: 4060RTX 6GB vRAM

Unfortunately, it’s very rare to find a gaming laptop with no dedicated GPU, thats why they’re called gaming.

Nonetheless the addition of this GPU does not increase the price significantly, it’s still around 1300 dollars or so as opposed to 2k or 3k dollars for a workstation laptop that has the same CPU & same amount of DDR5 RAM support.

If running 3D software, this GPU is going to be overkill. However, if you’re also a gamer, this worksout nicely because the 4060RTX here is running at 115W and that itself means it’ll be faster than a 3060RTX and have the same or even higher performance than a 3070RTX!

ASUS   ROG Strix Scar 16
PROS CONS
  • Latest 13th gen Core i7 CPU
  • 20 Threads (~40 VMs  )
  • 64GB RAM already upgraded
  • Latest RAM generation (DDR5)
  • 2TB SSD out of the box
  • large FHD display
  • Very Heavy
  • Very Low battery
  • Unncesessary GPU 

3. Lenovo ThinkPad E16 Gen 1

Budget Lenovo Laptop For Virtualization

  Intel i5-1335U 2P cores + 8E Cores (12 threads) 3.4-4.6GHz  

  16GB RAM DDR4 (Up to 40GB)

  Intel Iris Xe graphics

  512GB PCIe NVMe SSD

  15.6″ Full HD IPS Anti-glare

  4 lbs

  10 hours

  Windows 10 Pro 

  x1 USB 2.0 x 1 HDMI x1 Thunderbolt 4 (Type-C) x1  USB 3.2 Type-C

Okay, so far we have gone over the two most powerful and consequently expensive choices. The two laptops above are for extreme situations of 30+ VMs for the purposes mentioned (mostly labs & simulations).

We’ll now go over a much cheaper option (around 800 dollars) , this should be plenty of hardware to get you started in any field that requires you to run many VMs with medium loads on each.

Lenovo ThinkPad E Series: 

Within the lenovo thinkpads you have the E series, P series and the T series, the T/P series tend to be more expensive but more durable, resistant, better build quality and usually have significantly faster CPUs with more RAM support.

The E series are still high quality laptops bu they lack the same rock solid build design of the T series. You could say the build quality is just as good as most premium laptops , the T/P series are just a step higher.

CPU: i5 1335U (10 Cores)

CPU P-cores E-cores Cores Threads Virtual Machines (Max) E Speed P Speed
i5 1335U 6 8 14 20 40 3.6GHz 4.9GHz

They will usually come with a U (Low voltage) CPU which means it will draw less power and thus be restricted to lower clock speeds and this one of the reasons why the prices drop significantly. 

When it comes to virtualization though, high clock speed CPUs, are NOT needed unelss you’re running very high loads on VMs. Despite being ‘weaker’ than P or H series CPUs, it’s still 4.7GHz which is pretty high even by today’s standards.

Of course that clock speed is limited to the P cores which are reduced in Core i5 U CPU to TWO and the E cores (which are 8) run at 3.4 GHz.

So you have a total of 12 threads: 2 P cores (4 threads) and 8 E cores (8 threads). You can run 11 VMs (taken out the hosting operating system) efficiently with 4 of these running high load computations while the remaining  7 supporting a Windows OS with medium loads such as everyday tasks.

RAM: 16GB DDR4 (Out of 40GB. RAM)

RAM Slot 1 Slot 2 Total Max Windows VMs Linux VMs(1GB)
DDR5 16GB 16GB 32GB 48GB ~11(7) ~40

This E series model like most have one RAM soldered to the motherboard (in this case 8GB RAM) plus another 8GB RAM stick on the removable RAM slot making a total of 16GB RAM.

Now although the CPU (like all 13th generation CPUs) fully supports DDR5 RAM sticks, the manufacturer decided to use a DDR4 for the soldered RAM slot and that will limit the additional RAM Type & Speed you can add to the one that’s on the removable slot . In other words, you cannot use DDR5 on the additional RAM slot, you can only up the RAM to 40GB (using a 32GB stick) of DDR4.

Should you dare to add DDR5 , it will cause time delay issues and the computer may not work at all. 

Nonetheless, having 40GB RAM should be enough to run ~10VMs each with a Windows operating system on each and twice that many VMs each with Linux Distros.

Or you can run many VMs in ‘iddle mode’ or running code on each, this doesn’t take much RAM.

It should be around 2(Pcores)x2+8(Ecores)=12 threads x 2 VMs (iddle mode) = 24 VMs (-2 threads or one core for the host) ~ 22 virtual machines. Each can have approximately 2GB RAM.

If this laptop is out of your budget and you don’t need to run that many VMS but still want a Lenovo ThinkPad, check out the following cheaper options.

Some may have an older CPU however almost the same number of threads is available on each of these.

Link CPU Threads RAM Price
ThinkPad E16 Core i5-1335U 12 8GB (40 total) $679
ThinkPad E16 Ryzen 5 7530U 12 24GB (40 Total) $739
ThinkPad E16 Ryzen 5 7530U 12 8GB (40 Total) $561
ThinkPad E15 Core i5 1235U 12 16GB(40 Total) $699

 

Lenovo Thinkpad E16 Gen 1
PROS CONS
  • Latest Core i5 CPU
  • Supports 40GB
  • 12 threads (~24 VMs)
  • Okayish Core Count
  • Slightly heavier
  • Expensive if high storage
  • x2 256GB if 512GB (Must replace one to upgrade storage)

4. HP 17.3 

Cheap Laptop for Virtualization

  Ryzen 3 7320U 4 cores / 8 Threads 

  8GB LPDDR5 RAM

  ‎Intel UHD Graphics

  512GB SSD

  17” HD+

  4.5 lbs

  10 hours

    Wifi 6  

  Windows 11 S (Upgradeable to Windows 11 Home)

  x1 USB 2.0 x 1 HDMI x1 Thunderbolt 4 (Type-C) x1  USB 3.2 Type-C 

We will finish it off with the cheapest most common used laptop for 1-2VMs (Linux Distros or software testing).

This is what ALL newbies into virtualization need and by newbies I mean people who just run one or two VMs perhaps for running packages in a different OS like Linux or to test software on different OSs. 

CPU:  Ryzen 3 7320U

If you fall into that category, BELIEVE ME: any MODERN (post 2020) laptop will have enough CPU power to support an additional (or even TWO) Linux Distros running in the background PLENTY FAST.

Now with a budget under  500 dollars (350-450) you’ll find CPUs with either a Core i3 or a Ryzen 3. It’s important you pick the 12th , 13th gen Intel Core i3 or the latest 7th Ryzen 3 CPUs. 

Unfortunately, Intel Core i3 CPUs from the 13th generation are NOT currenly under 500 bucks, you may find them at 600 or even 700.

However, 7th gen Ryzen 3 CPUs are generally CHEAPER and they give you the same if not slightly BETTER performance when it comes to running VMs than the 13th Core i3.

This is because, as explained before, Ryzen CPUs have cores which ALL run close to the advertised clock speed as opposed to Intel CPUs which has the additional cores running at significantly lower clock speeds than the single core advertised clock speed.

This means there’s more CPU power available for each VM you run and each VM can run heavier loads than it would on an Intel Core CPU.

CPU Cores Threads Virtual Machines (Max) Speed
R3 7320U 4 8 16-2(host)=14 3.6GHz

How many VMs can we run?

For this Ryzen 3, that means 4 cores x 2 threads = a maximum of 14 VMs on iddle mode (total of 16 VMs but 2 threads for the host) and 6 VMs running medium loads (2 threads reserved for the host). Of course, you’ll be limited by RAM. 

Since you have 8GB RAM, you can only run 1VM of Windows (4GB) and 4-8 VMs of Linux (500MB-1GB VMs)

RAM: 8GB RAM DDR5

Another huge advantage of 7th generation processors over their Intel competitors is that they ALL support ONLY DDR5 RAM. That’s right, you won’t find DDR4 on a 7th gen Ryzen CPU

This is EXTREMELY beneficial for running heavy loads on VMs as the data transer speed on DDR5 memories is significantly higher than DDR4 which means the CPU can process data faster and say having a single core running at 3.9GHz on a Ryzen CPU will process data faster than a 3.9GHz on an Intel CPU simply because the difference in RAM memory.

Upgradeable to 16GB

8GB is plenty to run a Linux Distro on top of Windows. It’s even plenty for 2-3 Linux Distros in the background. You can also run an additional Windows VM with 8GB RAM.

Now…

If you’re running heavy data science algorithm for HEAVY data crunching on those 1-3 VMs then definitely upgrading it to 16GB RAM is going to speed up the process. If not, 8GB should be plenty for all other tasks running on a single additional VM such as programming on a Linux Distro with a Windows hypervisor.

Display: 17 inch HD+

Do note that this laptop has a HD+ display and not a FHD display! This is the reason why the price is significantly lower than laptops with the same hardware and size.

Luckily, it is a 17 inch display and if this is a laptop you’re going to be using mostly at the office or at home, then it can make up for the lack of resolution by giving you extra screen space. 

HP 17
PROS CONS
  • Latest Ryzen 3 CPU
  • Better Multicore performance than Core i3
  • 512GB Storage out of the box
  • Long battery
  • Latest DDR5 RAM (Up to 16GB)
  • Low resolution display (HD+)
  • Heavy
  • Does not support 40-64GB

Q: What is the best MacBook For Virtualization ?

For Windows VMs

It depends on what OS you want to virtualize. If you want to run Windows Virtual Machines on a Mac, then you will need a MacBook with a x86 processor IF you want to run the x86 version of Windows. In other words, you need a MacBook with an Intel Chip.

However, if you don’t mind running the arm version of Windows, you can use any Mac.

The newer MacBooks run on the M1 & M2 chips which are developed by Apple and are not compatible for x86 operating systems but they are compatible with arm-based OS.

On the other hand, the older pre-2020 MacBooks can run Windows Virtual Machines. You just need to install a hypervisor software (fancy word for VM software) that has a Mac Version like VM Fusion.

For Linux or OSX (Apple) VMs

If you want to run more OSX or Linux VMs, then any of the MacBooks, whether it’s the older or newer versions will work. Just be sure to calculate how many VMs you’re going to run SIMULTANEOUSLY and according to that buy a MacBook with the appropiate RAM.

You CANNOT upgrade MacBook (you can but it’s extremely difficult) so you should buy a macBook with the size of RAM you think you’re going to need.

Ideally, to run medium loads or heavy code on each Linux/OSX VMs, you want to reserve 2GB RAM. Linux can actually take less (as low as 500MB).

Q: What about the M1 & M2 MacBooks, can they run virtual machines too? 

Yes, but like I mentioned because they’re not CPUs with the x86 architecture, they do not support Windows VMs.

You are limited to Linux & OSX VMs as the newer processors are arm-based CPUs (they are not supported by Windows and I hear no rumors of Windows 12 being any different).

Q: How many VMs can I run with each MacBook?

4GB MacBook Air (Old): 3-4 Linux VMs, 1 OSX VM. No Windows (RAM too low)
8GB MacBook Air(Old) or MacBook Pro (Old): one Windows VM, 6-8 Linux VMs, 3 OSX VMs.
16GB MacBook Pro (Old): 3 Windows VMs , 12-16 Linux VMs, 6 OSX VMs.

Basically it all comes down to RAM. Windows will take 4GB, OSX will take 2GB and Linux can take as low as 500MB.


How to Choose A Laptop For Virtualization

We’ll go  over all the computer jargon behind the Virtual Machines topics while at the same time talk about the kind of hardware you need . We’ll also how each  piece hardware contributes to the number of VMs you can run.

Keep in mind that being able to run one VM doesn’t guarantee it’s going to run smoothly , it all depends on the kind of scripts & software you run on each of these.

If you are NOT new to the virtualization businessand you know the basic VM jargon, skip to the section  “Hardware Virtualization Specs”. 

1. VM Terminology

Components of VM mechanics. Image Credit: Michael Taschler

HyperVisor: This is the program ( script) that hosts and runs VMs. It creates a virtual processor, memory and storage and tricks the operating system (which is installed in your VM software) that this is all a new computer. 

VMWare ESXi: The name for a type of hypervisor  by VMware (entreprise class type-1 in technical terms) which is basically an operating system with its own kernel and which can host and run virtual machines.  This hypervisor by itself it’s not the full product software you can use like Oracle VM Box & VMWare workstation to run your VMs though. 

Kernelthe program that’s in control of every piece of hardware in a computer.

KVM: (Kernel Virtual Machine). Name for the kernel on Linux OS. It makes your Linux Distro turn into a hypervisor (thus making it able to hold VMs). 

Domain Controller: This is where the server or computer controls what programs or entities get access to join the network and use its resources . Basically the network security pillar.

Virtualization Software (Ex: VMWare Workstation): common name for the full software package that lets you instal land run VMs.  As of 2023, you can run Mac OSX, Chrome OSX, Linux OS and Windows OS virtual machines on windows laptops.

VMware Fusion: the most popular virtualization software for Macs.

Firmware: common name for the tool which gives you access to change a few parameters of the hardware of every PC. Ex: of the  This is exactly what allows you switch the booting device (from the hard drisk drive to perhaps a CD-ROM) when you want to run a different OS or install a new OS.

Another used term for it is “BIOS”. You usually see either of these or both names right after you boot up your machine. 

VPN:  This term has NOTHING to do with VMs, it stands for Virtual Private Networks. It’s what basically people use to get access to a server (or website) that’s restricted access to some regions.  The name is also used for software that will let you access your own computer from any other computer. Ex: Anydesk.

Why is the term useful here then ? 

PenTesting: because it is a common term thrown around VM forums.

Also, it is a term widely used by penn testers especially those running Virtual machines when they want to simulate a VPN or any other type of network for security testing which at the end all have the same goal: test the weak and strong points in a before launch network. 

VT:   terminology used by Intel to describe the technology embedded in their processors to run VMs. They’re just special instructions (similar to hyper threading) which makes the processor more efficient when running VMs.  Virtually every CPU post 2006 will come with VT instructions set but if you’re buying  or using old hardware you found on a basement, you can check whether it is “VT” capable or not in this Intel website Page

Hyper-V: A native hypervisor natively installed in most modern machines. If it’s activated, no VM software will be able to run. Thus it must be de-activated (it usually is by default) before you run VM software.

If you find out you’re getting errors when launching VM software, it is probably due to Hyper-V being enabled. When it is enabled, hyper-V will occupy a space called “VT-x” and this is the space that Virtualizatyion software use to run. You can read more about it  here.

This is because if VT-x is enabled (whatever that is), hyper V will “occupy” VT-x.


2. Virtualization FAQ

Now that you’ve got a good grasp of the jagon used in virtualization discussions, let’s clear a few misconceptions:

A) Is a core or thread limited to run one Virtal machine? 

If you do some research, you’ll find out most people will claim that each VM needs one FREE core to be able to run thus you cannot use that same core to run another VM.

This is not true…

When assining cores to VMs, hypervisors do not restrict one VM to one core or even one thread. CPU resources are balanced across all the virtual machines running. Now if you do ‘dedicate’ one core to one virtual machine, then you’ll get the best performance (in terms of processing speed) out of that VM, so ideally if you’re running some moderately hardware demanding task on each VM you want one core or one thread per VM. 

But say for penn testing purposes, most VMs just sit there in iddle with a few scripts running in the background and thus they will run fast even if a core is not reserved for a single VM. 

Q: Will assigning to many cores to a VM slow down performance?

Most likely it will but it depends on the software and what version of the software you’re talking about. Recent updates and version will handle it better.

Why did it happen? or why does it happen?

The more CPUs you assign to a guest (or VM), the more overhead there will be, more overheads means more time reading and processing data thus the longer it takes to process resources for a VM. You can avoid this problem by making sure you don’t add more cores to a guest than you think you will need, more cores won’t necessarily make it run ‘faster’. Your VM software will automatically assign the number of cores more efficiently without compromising processing speed.

Q: is it wise to buy the laptop with the best and latest hardware if I want to run many many VMs?

As you’ll see soon (in the last section) this isn’t true, if you do this you might be wasting too many resources on the GPU rather than RAM for example.

See, as i’ve said several times throughout this post, it’ll be very rare for you to be bottlenecked by CPU power as most CPUs (below 1000 dollars) on laptops will have 8 cores (the maximum amount as of 2023) and thus it’ll be down to how much RAM that laptop supports that will dictate how many virtual machines you can run simultaneously. After you max out on RAM & # Cores, you don’t want to maximize GPU power because that’s pretty much USELESS to run VMs and which is unfortunately what’s being ‘maxed out’ on laptops after CPU by manufacturers. You want to maximize clock speed and then storage instead.

Here’s the full hierarchy:

CPU Cores > RAM > clock speed > Storage in that order.

Q: Is old hardware better for Virtualization? 

You may have heard this somewhere and the reason is some people back then used to say that slightly older hardware will more than likely have the right set of drivers to run Linux with zero issues.

But today, in 2023, Linux distros has been updated left and right and constantly doing so thus most linux distros will now have the right set of drivers for most commodity machines. Linux is still Open Source and new revisions, updates are coming out every week.

So don’t be afraid of buying the latest 12th generation Intel Core CPU or the 6th generation Ryzen Chips, most distros will fully support laptops and computers with this hardware by mid-2023. 

Q: Is Intel hardware better for Virtualization?

It  was in the past but now both AMD & Intel have the same efficienty when running Virtual machines as similar technologies have been implemented in both brands to run VMs better. 

You should only pick Intel however if you’re on a budget as they may have a significant advantage when running VMs (due to the Vt-T technology being better implemented on older chips than the AMD counterparts at the time). 


3. Virtualization User Guide

This section will briefly talk about the activities and hardware you’ll need depending on the type of user you are. You can call it a complete wrap up of this post, I will talk about more hardware details in the last section but this may be all you need to read if you are not too adept with computer hardware. 

Students

If you’re in college or in an institution  taking classes say for the IT field  then you don’t need any special machine for the kind of VM work you’ll go through. Although you will run a few VMs for some classes, there’s no way you’ll be running dozens of VMs even if you take classes on Cybersecurity. Thus a commodity machine like a 350 dollar laptop made within the past few years (ideally with an 11th or 12th gen CPU) that has 8GB RAM will be all you need.

Software testing

If you’re a programmar and you’ve suddenly been told you need to work in the software testing department of your company then you ALSO don’t need to worry much about hardware to test software on different Operating Systems plataforms through Virtual Machines. 8GB RAM + A modern CPU will let you test software with zero lag (unless we’re talking about game development or graphically intensive software).

You don’t even need to worry about hardware compatibility issues as most VM software can run OSX, Linux, Chrome OS and Windows on most laptops. 

 IT professionals

I have a whole separate post on this topic here best laptop for IT Professional because it isn’t as simple as recommending this or that laptop. It really depends on what you’re working within the IT business.

For for those IT people that want to run several VMs, for like, a lab that integrates Cisco Firewalls,routers, switches with ESXi hosts you probably want to read the last section (hardware for virtualization) as you will need as much hardware goodness as you can get out of a laptop but to make it short and sweet most will be fine with 32GB RAM and an 6-8 core CPU.

Pen-Testing & Ethical Hacking

Most pen testers will at some point have to simulate large nested networks and they usually hold several dozens of VMs (in order to simulate a real world network). In this scenario, you’ll need to learn as much as possible about hardware for virtualization, thus it is almost a requirement to read the last section. I would in this scenario advice to buy a desktop as they are all around upgradeable.

Your CPU does not have too many cores? Change it. Need more RAM? Add more. This isn’t the case for laptops !

Buying the latest CPU generation is a good idea…

Newer CPUs have the advantage of being more and more optimized to simultaneously run a large number of VMs with several different operating systems more effciently.

Take as an example, running several Windows 10 VMs and a Kali VM. The Windows VMs are basically internal services while the Kali VM will act as a responder for poison requests. This simulation will be slightly faster with newer generation CPU as the instructions may be slightly more and more complex for older generations and the older you go, the higher the chances these executions will be cut off completely.

Another reason to max out on hardware resources…

Is to simulate a very realistic network (as the one found on companies) which will consist of a large number of VMs in order to test vulnerability points in it.

This means:

  • Making sure you get a laptop or motherboard (in the case of desktops) that supports RAM upgrades up to 32 or even 64.
  • A very spacious SSD (PCIe NVMe) if desktop as laptops are usually equipped with PCIe NVMe anyways.
  • The largest number of cores you can afford. Right now we have the 12th generation CPUs which work WONDERS for virtualization as the E-cores (additional cores) can also be used to run VMs (they’re equivalent to 6th gen Intel Core CPUs). Although Ryzen and Xeon will work as well depending on which one you can afford .
  • Probably wise to make sure your WiFi Card has the latest protocol ( 802.11ax) which is marketed as WiFi 6 as this allows you back-ward compatibility running scripts with older protocols which you want to do as ‘sniffing’ in network traffic comes in all different forms.
    • It is best to pick brands (and if possible) models that are more compatible with Kali Linux (Dell XPS or Lenovo ThinkPads come to mind) for these kind of tests.
  • A dedicated GPU, may come in handy to test scripts that make use of parallel processing to break in network security.

4. Hardware For Virtualization

 

CPU

Q: Which Processor is best for Virtualization ?

The CPU with more cores will always be the best one for virtualization. Even a much lower-clocked and older CPU will be better than a modern 2-core CPU even if you’re running only one virtual machine as it is likely you’ll also be multitasking while doing so . 

In an ideal world (unlimited budget) you would want to assign one core to each logical machine. All you can do for now (at least on laptops) is assign one thread to each VM and if running more than 15 VMs and you have maxed out # cores you can afford, then the next step would be (after maxing out RAM) to add more clock speed. 

Intel vs AMD for virtualization: Intel VT, Hyper V, AMD V

These are the two names for the technology on each CPU brand that makes virtualization more efficient. They both work just as well.

Note that netbooks (10-11 inch laptops), laptops with Celeron, Pentium , Atom or any other ‘mobile-friendly’ CPU may not have these features (VT-x, AMD-V, Intel Hyper V) thus they must be avoided even if you want to run 1 virtual machine. 

If you’re building a desktop and you’re browsing around eBay for older (thus much cheaper) hardware be sure to double check (through intel’s website or just the google search bar) they support:

  • Intel Vt, Hyper V, AMD V: Names for the Virtualization technology for both Intel and AMD
  • x86 or x64 plataforms: This means the CPU can run 86 and 64 bit systems (depending on the OS you want to virtualize)

Most modern CPUs will have all of that, it’s only a concern for much much older CPUs. Note that these features may have been disabled during the manufacturing process and you may have to manually go into the “BIOS/UEFI” settings to enable these.

Hyperthreading

Everybody knows what HyperThreading is, if you don’t please check my posts on the side bar for the basics.

It’s basically the programming paradigm responsible in splitting a “CORE” into two “cores”, the resultant cores are obviously “virtual cores” the same way a hypervisor fools the operating system there is a new computer within a computer, hyperthreading fools the system to think there are “TWO cores” instead of one.

Overall this multiplies the number of cores by two. Thus if you have a dual core CPU, you’ll have 4 virtual cores (also known as threads). You can use 1 thread for the hypervisor (host) and the other 3 to run 3VMs.

Now, you don’t have to check whether the CPU you’re buying or your laptop supports hyperthreading. All CPUs released after 2010 will 100% have hyperthreading (even the celeron and pentium CPUs). This is only an issue for older CPUs and I’ve mentioned here for information purposes.

12th & 13th Intel CPUs – E cores & P cores

The only exception to the hyperthreading feature being available in all modern CPUs are the 12th generation Intel CPUs.

It doesn’t necessarily reduce the number of cores or the number of threads but I just want you to know that a 12th or 13th generation CPU  will give you more or less the same # of threads as its 11th generation counterpart.

# Threads of the Core i7 1260P CPU as shown in Intel’s official page.

Why?

Now although there are two types of cores “E-cores” and “P-cores” so they seem to be more cores….and they make a total of 12, hyperthreading is only available on the “P-cores), thus despite having 12 cores, when hyperthreading is enabled, it will give you a total of 16 threads.

#Clock Speed = more VMs

Lastly, after you’ve found a bunch of laptops say all with 32 or 64GB and 16 threads, the next thing you can do to make sure MORE VMs can be run without lag or speed up the processing speed on each VM is to pick the CPU with the highest clock speed.

CPU resource control for virtual machines.

This may not be an obvious concept to grasp but if you look at the figure above, a virtual machines executions can be lined up in a queue where other virtual machines are also waiting.

A CPU (core or thread) can process each execution from every virtual machine if they’re all put in a queue as shown above, thus the faster the clock speed of said core , the smoother the executions will be and this translates to running more virtual machines with no waiting times (lag).

RAM

I remember running 4VMs on a gaming laptop that had the latest core i7 CPU and 16GB in 2017 (the usual set up of the most expensive gaming laptops) then using the task manager to see how much resources were being used and then found out the CPU was BARELY above 20% while RAM was choking near 80%, in other words, almost entirely being used. 

That doesn’t mean each VM will take 3GB that depends on how much RAM you assign to it and how much RAM your OS and software use. On average Windows 10 or 11 will take 3.5GB to run just as fast on a separate machine but for testing purposes you should be okay with 2GB and Linux Distros taken even less (500MB in some cases).

Anyways, the point here is that the more RAM you have (even for a low number of VMs) the better.

RAM is the main bottleneck for Virtualization. Hypervisors can easily distribute CPU resources across several VMs but RAM has to be partitioned and specifically assigned to single VM

8GB vs 16GB vs 32GB

Virtually all laptops above 400 dollars will come with 8GB out of the box and as long as they’re ultra thin 2-1 notebooks, they can be upgradeable to 16GB.

Once you step into 600-700 and above territory, you will get laptops that can have RAM upgraded to 32GB. If it’s got a gaming GPU, it can certainly be upgraded to 32GB, if it doesn’t you must do the research as it may not be upgradeable to 32GB (just 16GB). 

Take for example, the lenovo thinkpads: Only the T & P series can be upgraded to 32GB RAM with some of the high-end T series being upgradeable to 64GB.

Lenovo Thinkpads with 32GB on-board: You’re going to find quite a lot of overpriced thinkpads that have 32GB. Just remember that doesn’t mean you can upgrade them to 64GB RAM, most of the time, it’s just a model that has been upgraded all the way to 32GB before being sold.

At the end of the day, there’s no way to know whether a laptop can be upgraded to 32 or 64GB unless it is specified by the manufacturer or unless a review specifically says so (has opened up the machine to check).

8GB RAM 

Most laptops have at least this much and this will be good for most people that only need to run up to 2 VMs at the same time (IT support, It administrators, programmers, software testing).

An ethical hacker or a someone getting started with pen-testing will also do fine with this much at least in the beginning as the basic starting set up only requires you to Kali Linux + Windows. 

16GB RAM

You’ll need at least 16GB RAM to run 4 or more VMs 4 VMs.

And as long as you assign 1GB (Linux VMs) you can +10 VMs. 

I would ideally choose this much for both IT purposes & Penn testing. Here’s the typical set up for a common IT lab: 

ESXi Host 1 – 4GB of RAM – 20GB of Hard drive
ESXi Host 2 – 4GB of RAM – 20GB of Hard drive
Windows Server 2012 R2 installation – 15 GB of hard drive
+ linked clone – Domain controller & iSCSI target – 1GB of RAM – 35 GB of hard drive
+ linked clone – vCenter Server 5.5 – 4GB of RAM – 35 GB of hard drive

Total ~16GB RAM.

32GB-64GB RAM

32GB RAM becomes bullet proof for IT purpoes. As for penn testing and ethical hacking, you may need to upgrade your laptop to 64GB in some cases. 

DDR4 vs DDR5 

DDR5 is now avialable on 12th, 13th gen Intel Core CPUs found on laptops.

They are universal on laptops that have a 7th gen Ryzen CPUs because they do not support DDR4 anymore. 

DDR5 on average will operate at 1.5x faster data transfer speeds than DDR4, this means the addition of. DDR5 to an optimized laptop for Virtualization will boost VM performance significantly. 

But beware that NOT ALL 12th, 13th gen Intel Core CPUs offer DDR5, although they support that much RAM, if the manufacturer has used DDR4 on the soldered part, you are stuck with DDR4 because you cannot mix the two, it creates a sort of compatibility issues.

Storage

After you maxed out on RAM , # Cores and clock speed. You can speed up each Virtual Machine and maybe even squeeze a few more virtual machines through a high speed storage device. 

Solid State Drive vs Hard Disk Drive

Virtually every modern computer and laptop has now a solid state drive, there may be any outliers with a HDD but those are usually on either older models or very cheap laptops (under 300 dollars). 

That’s the first: make sure your laptop has an SSD. An SSD is up to x17 faster than HDDs!

Second Storage Drive Upgrade:

Most modern laptops also give you the option to add a second storage drive either through the M.2 slot or the 2.5” Bay.

If you want to maximize reading / writing speed , you want to do the upgrade through the M.2 slot because this is the slot where you can fit in a PCIE NVMe SSD as opposed to a SATA III SSD, where the former is x5 times faster.

Increase Storage Performance

Assuming you’ve done the upgrade, you can speed up performance by installing the Hypervisor (VM software) on the primary SSD (the SSD where the operating system’s instaleld) and use the secondary SSD to run VMs.  

In the off chance, you get a laptop that can support up to 3 SSDs, the more SSDs you have the more faster your virtual machines will be if you can evenly distribute their locations.

Ex: hypervisor on primary, 3 VMs on the second and 3VMs on the third for a total of 6 VMs.

External Storage

You can’t or rather you shouldn’t use an external storage as the host of additional Virtual machine data. However, you can use external storages to store all your VM data and hundreds of additional ISOs to free up enough space in the built-in SSDs on your laptop (as a good rule of thumb having an SSD used at 75% of its capacity will slow down performance slightly).

A much cheaper option would be the Adata HD720 line ($0.065/GB) .  This is the cheapest brand that will last outlast any other brand.  It is rugged, waterproof, drop proof, dust proof you name it!

Storage for Servers and VMs

Ex: A 3 node cluster (requires at least 5 VMs) with just one storage drive will cause I / O issues slowing down performance significantly. 

Solution: Using a RAID set up (or 3 storage drives)…

A Raid set up is basically having a bunch of storage devices installed within the same device. Likewise, you install the host OS on a single SSD and use the remaining SSDs to run all Virtual machines.

RAID Sets ups – Laptops

Raid set ups are rarely available on laptops. You’ll only found them on workstation laptop and only the expensive ones. It is rare for someone to build a server on a laptop but if you MUST do so, you can get either a workstation or try to use your current laptop’s CD/DVD drive port to install an additional storage. This is quite technical so only a comptuer technician, actually only a few, can pull this off.

In the worst case scenario, you can use USB-3 storage drives as additional storage (assuming of course you run out of space).

GPU (Dedicated)

Dedicated GPUs will not speed up performance on each virtual machine nor let you run more virtual machines. If you don’t know the difference between Dedicated and integrated graphics please check my post here.

Dedicated graphics (as opposed to integrated which come by default) have something like ‘extra cores’ which work the same way as CPU cores, however, they are only good either for graphical executions and simpler executions thus they are not useful for running virtual machines.

On the other hand…

If you’re running GPU sessions (apps & scripts) or parallel processing applications on your host machine (say for Media development or 3D applications) then they will definitely come in handy.  

Now , unfortunately, most high-end computers that come with the latest CPU with the highest number of cores AND 32-64GB RAM will most certainly (with a few exceptions) come with a dedicated GPU so sometimes you can’t avoid them.

Laptop vs Desktop

When does one become a better choice over the other?

Desktop

Most of the time, if you’re setting up a LAB / nested networks and running dozens of VMs on a single machine, it’s probably going to be a desktop. 

Advantages

  • Way easier to upgrade RAM
  • Way more storage
  • Way better CPU clock speed (though # cores usually remains the same as laptops)
  • RAID system support
  • Can handle more RAM (Up to 128GB on average)
  • Much much cheaper than laptops (with the same hardware).

Disadvantages

  • Desktops can be a lot louder.
  • Cannot bring them to a trip, thus it is impossible to have a mobile lab unless you use a VPN (AnyDesk for ex).
  • More hardware = more time spent doing maintance.
  • Can be expensive to upgrade.

Laptops

Only has two advantages over desktops:

  • Cheaper to upgrade
  • Mobile Lab

but several disadvantages:

  • It’s easy to run out of disk space with dozens of VMs.
  • Cannot run stuff in the background if you’re running dozens of VMs (due to lack of RAM or CPU resources).
  • It may not be upgradeable.

Comments?

If you have any suggestions, questions  or recommendations. Please leave a comment below.

Author Profile

Miguel Salas
Miguel Salas
I am physicist and electrical engineer. My knowledge in computer software and hardware stems for my years spent doing research in optics and photonics devices and running simulations through various programming languages. My goal was to work for the quantum computing research team at IBM but Im now working with Astrophysical Simulations through Python. Most of the science related posts are written by me, the rest have different authors but I edited the final versions to fit the site's format.

Miguel Salas

I am physicist and electrical engineer. My knowledge in computer software and hardware stems for my years spent doing research in optics and photonics devices and running simulations through various programming languages. My goal was to work for the quantum computing research team at IBM but Im now working with Astrophysical Simulations through Python. Most of the science related posts are written by me, the rest have different authors but I edited the final versions to fit the site's format.

3 thoughts on “4 Best Laptops For Virtualization 2024 (10-50 VMs ATSM)

  • Flumazenil

    Have you ever considered about adding a little bit more than just your articles? I mean, what you say is fundamental and all. However think of if you added some great visuals or video clips to give your posts more, “pop”! Your content is excellent but with pics and videos, this site could undeniably be one of the best in its field. Great blog!

    Reply
    • lol its two years later but thanks I will keep your suggestion in mind especially now that google is killing blogs with the new algorithm even though the content is great I need to do something about it.

      Reply
    • I’ve tried adding diagrams now and will soon try to add videos. Thanks for the suggestion I really need to change the site’s structure it seems google does not like it because I have lost a lot of traffic since september 2023.

      Reply

Leave a Reply

Your email address will not be published. Required fields are marked *