Basically, I'm confused after looking at so many Google Cloud products. I'm starting up a new project that includes a website, an iOS app, and an android app. I've decided to move forward with the Compute Engine as I'll have the flexiblility to do a lot stuff.
I'm thinking of using Cloud SQL for database service. I know that I can install MySQL on my VM. But I'm not sure what's the pros and cons. I'm still researching on this but in the mean time some experts opinion would be greatly appreciated.
TL;DR: Go with managed Cloud SQL. Better than doing it yourself and it doesn't cost much.
I'm no expert but I can tell you from previous experience that a managed database solution feels like much less of a hassle than doing it from scratch. Installing and configuring MySQL isn't especially hard, but it can get tedious (especially for devs like me who have done this many times over).
Also, when your app begins to grow, it'll just be a matter of pushing a few sliders to make your DB respond better to all the traffic. Trust me, you can enjoy a higher quality of life with words like "sharding" and "replication" not being part of your technical vocabulary.
Lastly, I don't remember Cloud SQL to be very expensive.
I have an interesting dev requirement that i've been trying to find a decent solution to for some time. I thought I might look here for suggestions.
I have 4 server locations on a network drive: 2 for development, 2 for production. each pair is intended to be identical, though because we're manually managing files gaps often surface. Developing in this context goes like this:
Make changes on one dev server.
Copy the updated files to the other dev server.
preview in browser.
repeat as necessary until ready to go live.
copy files from one dev server to one prod server, then the other.
review and test and hope to got you remember all the dependancies.
Realize that something is missing in a stylesheet somewhere because a colleague missed a server 6 months ago.
commence swearing. lather. rinse. repeat.
The closest thing to a solution I've been able to get to thus far involved using the local/remote server setup with autosave in dreamweaver with mappings to the two dev servers. But all this stuff being over a network makes things slow. And dreamweaver is far from my favorite editor.
I'd love to be able to find some way of making this better. In my freelance stuff I've been getting into things like NPM and Grunt, and I'm convinced I might be able to do something with something along the lines of task automation as is seen there, but I'm not experienced enough with these yet to know if that's a direction I should be considering.
Any suggestions would be greatly appreciated.
You can use any source control system you like to achieve what you want. Git seems to be the most popular choice those days, but even CVS would do the trick
Is it possible to mount Google Cloud Storage as a volume in a GCE instance for an application to access?
It is stil work in progress you can try s3fuse
Do not expect productional reliability just yet,
at least this was my experience about a month ago...
In addition to s3fuse mentioned in another answer, there are also gcsfs (based on s3fuse) and stormfs.
Note that I haven't tried any of them, so I can't speak as to how robust any of these solutions are.
Google has the official gcsfuse FS now. However, it's in the Beta and never will make it to stable release, since there are problems with the performance and stability which cannot be or are very hard to solve.
The performance degrades rapidly if you need to handle many small files. Also, scanning directories with many files (like find does) make take hours or days.
We have tried it for the production, but it has I/O errors occasionally. So now, we use it for the maintenance only.
I've been developing a web app locally on my local MAMP computer for the last few months. Now I am ready to launch it while continuing to add enhancements/fixes. So, I am wondering what is a good way to implement a development AND production server in order to efficiently manage updates, prevent overwrites, and seamlessly add other developers into the workflow. I also want something that has a minimal learning curve for me. Personally, for whatever reason, I've never been able to fully grasp version control systems like Git or SVN so I am hoping for an easier solution until I am able to invest more info the business.
As I see it, the options that I have are:
Spend more time learning Git before launching. And hoping that I don't break anything while further developing my app.
Buy two hosting accounts. One for Dev and one for Prod, where only I can do the deployments into Prod. I suppose I'd have to keep track of all files we've modified in a spreadsheet that are deemed ready for deployment.
Editing right on the FTP (no Dev server).
Are there any other options that you can recommend? I've heard that there are some new types of Web Hosting companies that can do the heavy lifting...
While personally, I have had good experiences using svn/git for multi-developer websites, I can understand your reticence to start relying on something you are not entirely familiar with. Unfortunately, I do believe that is your best option, but failing that, you might try using subdomains. My former employer would create test area on the disk and point beta.thedomainname.com at it. When bug fixes or upgrades were complete and verified to be working in the beta directory, the entire directory would be copied over to the live domain. Not the most elegant solution, but it worked. It certainly is cheaper than buying two hosting accounts.
Closed. This question is opinion-based. It is not currently accepting answers.
Want to improve this question? Update the question so it can be answered with facts and citations by editing this post.
Closed 7 years ago.
Improve this question
Duplicate
Reasons to Use a VM For Development
I'm trying to roll out a policy in my company where all developers have to work on a virtual machine (e.g. VMWare Workstations) that have the dev. environment such as IDE, tools, service packs already installed to make it easier for new team comers, smoother to provision new machines, etc...
do you recommend such an approach or do you work in a similar fashion in your company?
I've got a colleague who likes to work this way. He's got a virtual machine for each project he works on.
I personally don't like using a virtual machine to do development.
It's slower than working directly on
my machine.
It doesn't do multiple
monitors well.
Don't protect your devs from knowing the gritty details about IDEs, tools, and service packs. They need to know these things.
Also, don't force your devs to work a certain way. Some may not be happy about it, and unhappy devs = less productive devs.
I have worked with both methods for years. Currently I use VMs. They have many many advantages. However, don't force anyone into one particular way. They won't be productive if they are forced. If you can, convince them.
Advantages of VM for Dev:
Very quick deployment: One volunteer updates the image with the latest and customizes, and all get the benefit.
Each project can get a separate copy, no interference and no conflicts.
Very simple to "freeze" everything and restart! No need to save, close, run, load...
When things go wrong, it's an image, scape it, clone a new one and checkout your code.
Freeze while debugging or testing (sometimes you want to capture a specific state). Snapshots help if you want to go back and repeat some actions (think testing).
VMWare has remote debugging and backward execution!
Reproducibility! You devs and testers can reproduce bugs since the environment is controlled (assuming nothing other than work is on the image) and with saved states (assuming they use snapshots).
On the other hand, there are disadvantages:
VMs are bulkier, take a lot of space and memory.
You won't get 100% of your hardware performance.
You will lose some time on image maintenance.
Some people just hate it.
I highly recommend using virtual machines for development. Local virtual machines have very little performance penalty and make it much safer to try new ideas/software.
Just make sure you have enough RAM to allow for several VMs and the host OS.
See also
Where i work at the policy mandates that we all have a physical machine wich runs a VM. We only have admin privileges to the VM and not the physical machine. This tends to create problems when we have to run several development applications, builds tend to be slow, everything is slow for that fact. Also when the VM starts reaching the 15gb limit (around a month and a half use) things get complicated as the VMs start crashing and we need to ask for VM compression.
My experience has been bad so i wouldn't recommend it. We usually run the following applications in the VM : Text editors, IDE, Weblogic instance, TOAD for database access. Explorer and Firefox, office applications and less.
With modern IDEs there's a lot of graphics and disk IO going on, neither of which is performed well by VMs. So - if your VM responds fast enough for the developers to use, then I'd say there's no reason why not. If it doesn't you either need to get a faster machines for them or go back to documenting how to setup the build environment.
the other factor against VMs is that if you change the environment, you have to do it for all VMs, and document the changes anyway. If you're telling everyone how to set up their system, you might as well let them set their own system up on the base metal.
Incidentally, we do have VMsa for this - but they tend to be for old versions of the product, so we can still build it without having to install the old service packs, sdks and compilers. Its ok, but I find installing everything locally and switching between them (using junction to point to the build directories) is easier.
Now IIRC VMware has a virtualisation project called thinApp that transparently puts a OS environment onto your local box, so you can have several conflicting applications running side-by-side. I've not used it, but did look into it as something that might be better than whole guest VMs running in their own windows.
Personally, while I feel it's a good idea for all the reasons you mentioned, I also feel that it requires quite a bit of extra cost on machines. I was just trying out Windows 7 over the weekend on VMWare and a moderate machine (AMD X2 4600, 2 GB RAM) I find that working in a VM can very much be a worse experience than working on top of the real hardware.
At our shop, we pretty much use all VM's for development. One useful strategy we've employed though to increase VM performance, is to always run them on a high speed external hard drive. Doing this makes them run incredibly fast, since VM's usually a demand a lot of disk IO, on as the prior post mentioned.
There are valid reasons to use VMs for development. However, if you're thinking of doing this just to standardize development environments across your organization, there are better ways to accomplish that (ie, having standard machine images).
In some cases, like doing SharePoint dev work, you are more or less required to work on a server, and I just don't like the idea of turning my laptop into a 2003/2008 server :-)
We have two VMWare ESX boxes that hosts our dev machines and it works great as long as people remember to switch off those images that are not in use. Another advantage is that we have a complete network of ESX images in their own domaine which gives us the abillity to do a lot of funny stuff :-)
Start with some developers and try to gather some actual data about productivity change.