I'm in need of a new laptop. Could anyone please recommend some important specs that I should look out for when choosing one. I do have access to a HPCC, but would like to do some smaller analyses locally?
Hi Pierre. I will probably be mainly working with R on pre-calculated expression matrices that I can perform on the cluster. Although I must admit, having the option of working with raw data is tempting, even if it's just for a subset for workflow testing.
I do have access to a HPCC, but would like to do some smaller analyses
locally?
This may come down more to what is your budget and lifestyle need (e.g. weight of laptop, need to be able to take it with you every day on a commute etc.). Since you can move any of the heavy lifting to HPCC at any given time, something modest may work ($600-1000). If you are in a position to splurge, then there are plenty of high end options between $1500-2500 (Dell XPS, HP Spectre, Macbook Pro, MSFT Surfacebook).
HP Pavillion (purchased in London in 2015; ~£550 GBP)
16GB RAM
2 TB hard-disk
CPU is AMD A10-7300 Radeon R6, 10 Compute Cores 4C+6G (4 CPU cores +
6 GPU cores)
3 USB, 1 HDMI, ethernet port, SD port, WiFi, bluetooth
16 inch screen
64 bit
Linux (Ubuntu 16.04) (was originally Windows but removed it when I purchased it)
This is quite bulky, though, and takes up half carry-on luggage weight when flying. However, I can do virtually all analyses using it, including genome assembly of bacterial species. It is also now literally falling apart and is charred from overheating in hot climates.
This laptop is not suitable for carrying around - it is literally bulky and heavy, but powerful.
Lenovo ideapad 330S (purchased in Brazil in 2019; ~US$600)
1 TB hard-disk
8GB RAM
CPU is Intel i5
2 USB, 1 HDMI, SD port, WiFi
64 bit
Windows 10
This is lightweight and used for, e.g., teleconferences, MS Office, working while traveling, and permits most level of work via R. For example, I could easily process raw data on the bulky laptop (above) and then export an Rdata object for working on this. The battery lasts 5 hours when the screen is dimmed.
Thank you for the excellent feedback Kevin! Much appreciated. I will probably be mainly working with R on pre-calculated expression matrices that I can perform on the cluster. So I guess in that case I am looking more towards the lighter option. Although I must admit, having the option of working with raw data is tempting, even if it's just for a subset for workflow testing.
what kind of analyses ?
Hi Pierre. I will probably be mainly working with R on pre-calculated expression matrices that I can perform on the cluster. Although I must admit, having the option of working with raw data is tempting, even if it's just for a subset for workflow testing.
This may come down more to what is your budget and lifestyle need (e.g. weight of laptop, need to be able to take it with you every day on a commute etc.). Since you can move any of the heavy lifting to HPCC at any given time, something modest may work ($600-1000). If you are in a position to splurge, then there are plenty of high end options between $1500-2500 (Dell XPS, HP Spectre, Macbook Pro, MSFT Surfacebook).
Maybe one certified by Canonical in case you decide to dual boot: https://certification.ubuntu.com/
UPDATE: apparently a Raspberry Pi would suffice ;P https://www.sciencedirect.com/science/article/pii/S1874939919300409?via%3Dihub