Service and dependency inventory
For each server we document services, ports, databases, hardware-bound licenses and links to other systems.
Start here
Free Audits 5Urgent help
Emergency 4Ongoing support
Managed IT 5Projects
Servers & Hosting 6 Network & Virtualization 9 DevOps 5 Security & Recovery 5 Hardware & Licensing 2Find out where your servers, security, backups and performance stand, at no cost.
Server down, network out, site hacked or data lost? Call us now.
Monthly network and server support with response times written into the contract.
Setup, configuration, management and migration of Linux and Windows servers, panels and mail.
Network design and cabling, MikroTik, VoIP, branch links, virtualization and private cloud.
Networking & communications
Virtualization & cloud
Containers, Kubernetes, automated delivery, infrastructure as code and observability.
Server hardening, firewalls, backup and DR, ransomware recovery and incident forensics.
Advice, supply and installation of servers and network gear, plus genuine enterprise licenses.
Start here
Urgent help
Ongoing support
Projects
Old physical servers tend to be the ones nobody dares to touch. Our cloud migration work starts with a full inventory of services and dependencies, followed by a test run on a copy. The live service moves only after you sign off, and the old server stays powered off and untouched until the test period ends.
Cloud migration is moving servers and their services from old physical hardware or another virtualization platform onto virtual machines, a data center or cloud servers, with a tested plan and a way back. It is typically needed when a core application still runs on Windows Server 2008, a VMware license won't be renewed, or cloud servers were bought and nobody knows where to start. PikoSystem documents each server's services, ports, databases and hardware-bound licenses, converts physical machines with Disk2vhd, Clonezilla or virt-v2v, and moves VMware VMs to Proxmox with its import tool or qemu-img. Every server is trial-migrated on a copy first, and the old one stays untouched until the agreed test period ends. You receive a migration plan with rollback steps, a post-migration test report per service and destination documentation.
For each server we document services, ports, databases, hardware-bound licenses and links to other systems.
Windows and Linux servers are converted to VMs with tools such as Disk2vhd, Clonezilla or virt-v2v, and drivers are fixed afterwards.
VMs are moved with the built-in Proxmox import tool or qemu-img, VMware Tools are removed and VirtIO drivers installed.
We move VMs to cloud servers or colocation, set up addressing, DNS and firewall rules at the destination, and connect it securely to your office.
For each service we write down the downtime window, migration order and the path back to the old server.
The classic failure is a Windows VM that won't boot and shows INACCESSIBLE_BOOT_DEVICE, because its disk was attached to VirtIO SCSI before the driver was installed. Attach the disk as SATA first, install virtio-win, then switch the controller.
VMs that booted with UEFI on VMware need OVMF firmware on Proxmox. On Linux, the NIC name changes, for example from ens192 to ens18, and the static IP no longer applies. Windows sees a new adapter and leaves the IP settings on the old one. Since Proxmox VE 8.2, the built-in ESXi import tool handles much of this work.
Disk2vhd copies a running server using a VSS snapshot, but databases such as SQL Server are only crash-consistent that way. Stop the database service or take a separate database backup first.
On Linux, the physical server's initramfs usually lacks VirtIO modules and has to be rebuilt with dracut or update-initramfs before it boots in the new environment. A Windows OEM license bought with the hardware does not move to a VM, so proper licensing for the virtual environment is needed.
We generally don't P2V a domain controller. Building a new DC, transferring FSMO roles and demoting the old one is safer, because restoring an old copy of a DC can cause USN rollback and break replication.
Windows Server 2008 machines can be moved, but they no longer receive security updates. If the application can't run on a newer OS, keep the VM in an isolated VLAN with tight access. Applications that open files from an office share become very slow once the server sits in the cloud, so check the architecture before migrating.
Beyond CPU and RAM, cloud cost includes outbound traffic, backup storage, extra public IPs and a reliable office-to-cloud link. Once every user reaches services in the cloud, an office internet outage cuts access to everything, so a second link matters.
Keeping your only backup with the same provider is a risk too. Before the final cutover, lower DNS TTLs so IP changes take effect quickly, and check legal requirements on where data may be stored.
We review servers and services and document dependencies.
Migration order, downtime windows and rollback are agreed with you.
We copy each server, boot it at the destination and have key users test it.
Final data is synced and the service is switched over. The old server is kept until the agreed test period ends.
Downtime depends on disk size and network speed, and we estimate it per VM. Less critical VMs move during working hours, core ones at night or on a weekend. If something goes wrong, the original VM is powered back on in VMware.
Many USB dongles work inside a VM through USB passthrough or USB over IP. Licenses tied to hardware IDs may need reactivation by the vendor. We check these cases before the move.
It depends on the type of data, legal requirements, monthly cost and how much control you need over hardware. During discovery we lay out the options side by side with cost and risk, so the decision is well informed.
Related searches
Quote
A few lines on where things stand and what you want is enough. An engineer calls you back, not a sales rep.