I am looking for a way to master-slave two identical (hardware, os and disk image) Macbook Pros such that any interface actions made by the user on the master MBP with be sent to the slave MBP (via LAN or some other data connection): mouse movements, application is launched, user edits file, etc.

Similar to how one computer can control another via video chat applications like Citrix, but instead of having the master working inside a portal that controls the slave, in this case the slave would be repeating all the actions made on the master at a "desktop level."

Does anyone know of a currently available solution? If one doesn't exist, what are the major things to consider in terms of developing one. Obviously interface actions need to be streamed from the master to slave and then executed on the slave.

Any insights would be greatly appreciated, thanks!

  • It might help if you edit your question to tell what your ultimate goal is here, to avoid what we call the X Y Problem. – Spiff Jun 10 '17 at 6:20
  • Latency is gonna kill you here. Simplest possible scenario...Launch new app on A; accept EULA; Open file; work. B hasn't launched in time & misses the EULA click. – Tetsujin Jun 10 '17 at 7:08
  • Something like this maybe? amazon.com/IOGEAR-Cable-Switch-Audio-GCS72U/dp/B002K0TU2C – D Schlachter Jun 10 '17 at 20:01

If I understand you correctly, you want a physical mirror between two computers where the same activity is being simultaneously processed by both rather than one simply having a remote session to the other. It is not possible to do this.

Something similar to what you are talking about does exist in the world of failover clustering and live replication, but it's only done on high-end servers in corporate environments where high-availability and disaster recovery are required. Even in those scenarios, however, it doesn't work the way you are thinking. A failover cluster, a just one or more standby servers kept in a "warm" state so that any of them can immediately pick up the load where the other left off in the event one of them fails. But it is not a cooperative effort; only one server is actually in production any given time with the others being a warm standby state.

The other kind of cluster would be a high-performance compute cluster where a bunch of nodes collaborate on a particular task. This is how supercomputers work, but it requires specialized software and isn't something you'd do for general computing tasks.

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.