Explore our new Cybersecurity Terms Glossary, explained by our experts.
Cybersecurity Terms Glossary
Memory acquisition copies a running system’s RAM to a file before shutdown, preserving decrypted payloads, command lines, open sockets, and credentials that exist nowhere else on the machine.

How Do You Acquire a Memory Dump?
Acquiring a memory dump means copying the contents of a running system’s RAM into a file an examiner can parse later. It matters because a large share of the evidence in a modern intrusion exists nowhere else: the decrypted configuration of an implant, the full command line of a process launched minutes ago, the socket opened to a staging server, and the credentials still resident in LSASS.
Pull the power cable and that evidence is gone. Leave the machine running without capturing it and an attacker with hands on the keyboard may remove it for you. Acquisition is therefore usually the first technical action in an incident response engagement, taken before the host is isolated at the network layer wherever that sequence is possible.
RFC 3227, still the reference text for evidence collection, ranks sources by how quickly they decay: CPU registers and cache, then RAM, then network state and running processes, then temporary files, then disk. On a suspect Windows host that translates into a memory image first, a short set of live volatile data commands second, then disk imaging or triage collection.
One exception is worth planning for in advance. If the machine is mid-encryption during a ransomware attack, containment beats collection: isolate first, then capture whatever memory survives.
Paging artefacts matter too. hiberfil.sys, pagefile.sys, swapfile.sys, and crash dumps at C:\Windows\MEMORY.DMP hold compressed or paged-out fragments of earlier memory states, and they should be lifted from the disk image even when a live dump succeeded.
Virtualisation offers a cleaner route. Suspending or snapshotting a VMware guest writes a .vmem file that Volatility parses directly; Hyper-V produces .vmrs; a KVM guest yields an image through virsh dump --memory-only. None of these runs anything inside the guest, so the attacker sees no new process and no driver load.
In AWS or Azure the equivalent is to execute AVML through the instance management agent and write the output straight to object storage in a separate forensic account, then take the disk snapshot. Memory before snapshot, always, because stopping an instance destroys its RAM. The disk half of that workflow is covered in snapshot-based cloud forensics.
A truncated or unparsable image discovered three days later is a lost case. Confirm the file is usable while the host is still in front of you:
python vol.py -f memdump.raw windows.info
python vol.py -f memdump.raw windows.pslist
If the kernel base and build number resolve and the process list looks sane, the acquisition worked. If they do not, repeat with a different tool before releasing the machine. At Privia Security, memory capture is a standard first step in our Investigation and Incident Response Services, because the evidence it preserves often decides whether an intrusion can be explained at all.
Capture memory first where it is safe to do so. Powering off destroys every running process, open connection, and decryption key held in RAM, and those are frequently the only proof of what the attacker did. The exception is active destruction: if ransomware is encrypting files or data is leaving the network, isolate the host immediately and accept the loss of memory evidence.
Plan for the full size of installed RAM. A laptop with 16 GB produces a 16 GB raw file; a virtualisation host with 512 GB produces 512 GB. Compressed formats such as AFF4 reduce storage on systems with large amounts of zeroed memory, but you should still provision destination space at full capacity so the capture never fails part way through.
Yes, unavoidably. Loading a driver and allocating a buffer changes memory contents, which is why memory evidence is treated as a reasoned reconstruction rather than a perfect copy. The accepted practice is to use a tool with a small, well-documented footprint, record exactly what was run and when, and be ready to explain the change in a witness statement.
Yes, and it is the preferred method. Suspending a VMware guest writes a .vmem file containing guest RAM, Hyper-V produces a .vmrs file, and libvirt offers virsh dump with the memory-only option. Because the capture happens at the hypervisor, no new process appears inside the guest and the attacker has no opportunity to react to the collection.
You May Be Interested In These
We use a few cookies that are needed to make this website work, and they are always on. We would also like to set analytics cookies, to see which pages are useful; functional cookies, so that our live chat works; and advertising cookies, to measure our campaigns. We will not set any of those until you say yes, and you can change your mind at any time from the Cookie Settings link in the footer.
Choose the cookies you are happy for us to use. Strictly necessary cookies are always on; everything else is off until you turn it on.
These cookies keep the website working: they hold your security checks, your form submissions and the choice you make here. The site cannot run without them, so they are always on.
These let our live chat open and remember your conversation, and let the newsletter sign-up work. Switch them off and the chat window will not appear.
These tell us which pages are read and how people find us, so that we can improve the site. The reports are about groups of visitors, never about you personally.
These let us measure our advertising and show you our adverts on other websites, such as LinkedIn, Google and Meta. Switch them off and you will still see adverts, but they will be less relevant.