HWZ Forums

Login Register FAQ Mark Forums Read

WEB HOSTING COMPANY TO RECOMMEND!

Like Tree83Likes
Reply
 
LinkBack Thread Tools
Old 16-04-2013, 04:47 PM   #31
High Supremacy Member
 
Join Date: Nov 2010
Posts: 33,628
VPS gives you root access, meaning you can install anything you want (include Rails), and you have to install them yourself.

I think the decision factors are: what you want and what you know. If you are not familiar with sysadmin and have no time to learn it then PaaS like Heroku will be a better solution for you. If want custom software and configuration and you know how to do it, then VPS will be a better solution for you.
oh.. i wanna try to host my site with RoR because it's easier, as compared to PHP or ASP. (or maybe it's not? haha...) but from what you say, i think the steps of setting up the RoR environment seem tedious, if the web hosting company decided not to do it for me. then i might as well ask them to give apache and i just stick to PHP. haha.

Go try AWS
okay, i will check it out. thanks!
NSforSG is offline   Reply With Quote
Old 16-04-2013, 04:48 PM   #32
High Supremacy Member
 
Join Date: Nov 2010
Posts: 33,628
NSforSG might not know what VPS is. So AWS might be too complicated for him/her. Shared hosting or PaaS solutions like Heroku will be easier for him/her.
is AWS the same as OpenShift?
NSforSG is offline   Reply With Quote
Old 16-04-2013, 05:10 PM   #33
Junior Member
 
expertleong's Avatar
 
Join Date: Apr 2006
Posts: 25
Both Openshift and AWS offer could services. Openshift is operated by Redhat, AWS is offered by Amazon.

You might be looking at too many thing at the same time, that will confuse you. First decide on the programming language first. If you are already a programmer, Ruby/PHP/.NET are equally easy to learn. Just pick one language and settle for a shared hosting (e.g. AppFog, Heroku, GAE) first. Concentrate on building awesome apps, leave the hosting and sysadmin to the experts.
__________________
Leong Hean Hong
Mobile & Web Developer
http://about.me/hongster
expertleong is offline   Reply With Quote
Old 16-04-2013, 05:15 PM   #34
High Supremacy Member
 
Join Date: Nov 2010
Posts: 33,628
Both Openshift and AWS offer could services. Openshift is operated by Redhat, AWS is offered by Amazon.

You might be looking at too many thing at the same time, that will confuse you. First decide on the programming language first. If you are already a programmer, Ruby/PHP/.NET are equally easy to learn. Just pick one language and settle for a shared hosting (e.g. AppFog, Heroku, GAE) first. Concentrate on building awesome apps, leave the hosting and sysadmin to the experts.
thanks for your advice! i did a website @ openshift, i like their free hosting. but the premium can be crazy if i need more data storage. that's why i'm thinking of moving away from openshift.

since AWS = openshift, i may try AWS. and like what you advise, i will leave the hosting and sysadmin to the experts. haha.
NSforSG is offline   Reply With Quote
Old 16-04-2013, 05:20 PM   #35
Junior Member
 
expertleong's Avatar
 
Join Date: Apr 2006
Posts: 25
AWS EC2 requires you to setup Linux and install the necessary software. I am suggesting you avoid AWS and go for services like AppFog/Heroku/GAE.
__________________
Leong Hean Hong
Mobile & Web Developer
http://about.me/hongster
expertleong is offline   Reply With Quote
Old 16-04-2013, 08:27 PM   #36
Senior Member
 
masterstroller's Avatar
 
Join Date: Dec 2011
Posts: 786
Been using godaady vps for 3 years... Cheapest ive found so far.. Stick with the windows for easier upkeep. The servers seem to get reset like once in 3 months. For windows not much effect. For linux it clears the /etc/Resolv.conf deleting your dns server list.. Preventing your mailing service from sending out mails.
masterstroller is offline   Reply With Quote
Old 16-04-2013, 09:42 PM   #37
Arch-Supremacy Member
 
davidktw's Avatar
 
Join Date: Apr 2010
Posts: 10,044
AWS EC2 requires you to setup Linux and install the necessary software. I am suggesting you avoid AWS and go for services like AppFog/Heroku/GAE.
Have you look at Elastic Beanstalk ? However if it is about ROR then it's not found natively in beanstalk, if really it's required, nothing is stopping you from having one in AWS. Look at http://bitnami.com/stack/ruby

However, curiously, or in fact, it's a pity wondering when it becomes installing linux and applications in it becomes a system engineer job or the infra team job, and one actually need an expert for such stuffs ? So a software developer shouldn't be proficient in such tasks ? No wonder the IT standard in this part of the world is going downhill....

Last edited by davidktw; 16-04-2013 at 10:17 PM..
davidktw is offline   Reply With Quote
Old 16-04-2013, 09:48 PM   #38
Junior Member
 
Join Date: Mar 2010
Posts: 89
one thing really bad about AWS and GAE is vendor locking -.-". Especially true for GAE
stupidbodo is offline   Reply With Quote
Old 16-04-2013, 10:11 PM   #39
Arch-Supremacy Member
 
davidktw's Avatar
 
Join Date: Apr 2010
Posts: 10,044
one thing really bad about AWS and GAE is vendor locking -.-". Especially true for GAE
Good question about the lock in. Perhaps it's a good debate, what would your recommendation be ?
davidktw is offline   Reply With Quote
Old 16-04-2013, 10:58 PM   #40
Junior Member
 
Join Date: Mar 2010
Posts: 89
I think overall it depends on what you are trying to achieve. I see many misconception that you only need to know how to write features to build a web app, unfortunately, that is not the case.

Writing a web app is easy, but writing a web app that can scale efficiently is really hard.

Suppose a new developer comes in and wants to build a new groupon clone. He picked up lessons from codecademy and begins writing his app. Few months later he finished his app and deploys it.

What is misleading is that many people thinks that once they finished writing the app they are done. Once your site turns operational and hot that's where the trouble comes in. Queries that normally takes sub milli seconds to work on your pc is now running minutes on your site.

How you write your code(whether is it spaghetti or course meal), how you deploy it(how many clusters, how many database servers, how your queries get passed) all these matters alot if you want to go big.

It is best you think of deployment in long term. I give you a very good scenario if you are going hosted and your site gets so awesome that you want to do some data analysis from your users, you decided to deploy a data warehousing system. You begin looking at options and found this bloody cheap and awesome AWS Redshift that costs only $1000/yr/tb (Amazon Redshift) . But when you read the faq, you realized that
in order to import your data you need to either use S3 to do it free or sent it via data pipeline. If you intend to use data pipeline this deal doesn't seem so great anymore as you will need to incur additional expense.

Every vendor has their pros and cons, I don't support any because I use it based on what I am trying to do. Heroku is good because it is easy to deploy and once you scale out you don't have to waste so much time thinking about how to configure multiple servers. Some people might say its expensive but depending on your case, you might saved it on labor costs.


I am a wheel dealer, what I do is I get a dirt cheap hourly priced hosted or cloud service like AWS for development (why aren't you using that? heck you can do all sorts of funny things).

BUT when it comes to real deployment, I make sure I planned carefully so that I don't incur so much technical debt (Technical debt - Wikipedia, the free encyclopedia) .

Unfortunately if you want to build a successful app, you have to at least understand what is the best way to deploy your site. Okay, back to writing code.

Last edited by stupidbodo; 16-04-2013 at 11:13 PM..
stupidbodo is offline   Reply With Quote
Old 16-04-2013, 11:00 PM   #41
Junior Member
 
Join Date: Mar 2010
Posts: 89
Have you look at Elastic Beanstalk ? However if it is about ROR then it's not found natively in beanstalk, if really it's required, nothing is stopping you from having one in AWS. Look at Ruby Stack Cloud Hosting, Installers and Virtual Machines.

However, curiously, or in fact, it's a pity wondering when it becomes installing linux and applications in it becomes a system engineer job or the infra team job, and one actually need an expert for such stuffs ? So a software developer shouldn't be proficient in such tasks ? No wonder the IT standard in this part of the world is going downhill....

I like the part on "No wonder the IT standard in this part of the world is going downhill...."
stupidbodo is offline   Reply With Quote
Old 16-04-2013, 11:49 PM   #42
Junior Member
 
expertleong's Avatar
 
Join Date: Apr 2006
Posts: 25
It depends on needs of individuals/company. I am not discouraging all developers to avoid basic sysadmin skills. The suggestion was made based on understanding of NSforSG's situation. He has not made up his mind on what programming language to learn, I concluded that he is still/will be learning programming language. He is not sure about various hosting options and unfamiliar with terms like "VPS". It sounds like he is new to sysadmin. I suggest he concentrate on learn programming and developing the web app first at this stage. In future when he needs more sophisticated configuration or need to scale/optimize his app, then he can start looking into solutions like AWS/Openshift.

It is a plus, if a programmer has sysadmin knowledge. But I do not think this is very essential skill to have, if the company already has a dedicated sysadmin. Sometime there is a need for specialization in certain field, there are also time where there is a need for a person to be jack of all trades.

"No wonder the IT standard in this part of the world is going downhill...." I do have similar observation on Singapore's software developers. I think this is mainly due to lack of interest in software development. Programmers don't get much respect here. And most of the Computer Science graduates are not really passionate about programming in the first place, programming is just a job to many programmers here. I can understand why Software Development field is going downhill. This article pretty sums it up quite well: Rise of the Software Craftsmen - SGE : SGE
__________________
Leong Hean Hong
Mobile & Web Developer
http://about.me/hongster
expertleong is offline   Reply With Quote
Old 17-04-2013, 12:53 AM   #43
Arch-Supremacy Member
 
davidktw's Avatar
 
Join Date: Apr 2010
Posts: 10,044
It depends on needs of individuals/company. I am not discouraging all developers to avoid basic sysadmin skills. The suggestion was made based on understanding of NSforSG's situation. He has not made up his mind on what programming language to learn, I concluded that he is still/will be learning programming language. He is not sure about various hosting options and unfamiliar with terms like "VPS". It sounds like he is new to sysadmin. I suggest he concentrate on learn programming and developing the web app first at this stage. In future when he needs more sophisticated configuration or need to scale/optimize his app, then he can start looking into solutions like AWS/Openshift.
The approach while feasible, in my opinion is not ideal. As much as today we are talking about using the cloud, it doesn't really mean things must start halfway from the cloud. The knowledge to develop properly in the cloud boils down to how deep one understands the fundamentals that trace all the way back into the soil.

As you have mentioned, TS haven't decided on which programming language to approach, which certainly means it's a start. How far can development goes without understanding the platform that one is riding on ? It's like stepping on someone's shoulder and keep climbing, eventually when the bottom layer starts to alter, the top just crumble with nothing to hold on to.

That's why my perception to cloud development is, because one has the fundamentals to properly develop from ground up that makes riding on the cloud a swift. Without mentioning which platform one starts development on, it goes without saying one need to understand how a system runs, and where the code executes, how it's run.

If it's a proper advice to a new comer, it should be always start low and move up the chain. If one can't command the underlying platform directly below it, one have not earn the rights to move up the chain. If one can't even understand how to setup the platform to execute the codes, then one better put it as first priority. If not, there is nothing about scaling and optimisation that matters. It's the foundation that matters.

It is a plus, if a programmer has sysadmin knowledge. But I do not think this is very essential skill to have, if the company already has a dedicated sysadmin. Sometime there is a need for specialization in certain field, there are also time where there is a need for a person to be jack of all trades.
I really don't think it's a "plus" that a programmer has the so called "sysadmin" knowledge. How is operating an operating system just a sysadmin knowledge ? I'm puzzled. Everyone need a system to work on, so suddenly knowing how to operate in your own laptop or desktop, which could be either Windows, Mac, Linux becomes a system administrator role ? If it is indeed a system administration role, then it better be one of the criteria to put into the resume of a developer because I don't understand how can a developer earn the rights to be known as one without the skill set of managing the very layer that make things tickle.

I do understand where you are coming from when proposing one doesn't need to be jack of all trades, I believe somewhat in that too should one doesn't want to earn the title of being versatile. However I probably will place those under the differentiating between a BI expert, DB expert, security expert etc as been different verticals. I hardly call knowing how to manage the operating system an expert role. It shouldn't even be a vertical by itself.

Last edited by davidktw; 17-04-2013 at 01:16 AM..
davidktw is offline   Reply With Quote
Old 17-04-2013, 01:12 AM   #44
Arch-Supremacy Member
 
davidktw's Avatar
 
Join Date: Apr 2010
Posts: 10,044
I think overall it depends on what you are trying to achieve. I see many misconception that you only need to know how to write features to build a web app, unfortunately, that is not the case.

Writing a web app is easy, but writing a web app that can scale efficiently is really hard.

Suppose a new developer comes in and wants to build a new groupon clone. He picked up lessons from codecademy and begins writing his app. Few months later he finished his app and deploys it.

What is misleading is that many people thinks that once they finished writing the app they are done. Once your site turns operational and hot that's where the trouble comes in. Queries that normally takes sub milli seconds to work on your pc is now running minutes on your site.

How you write your code(whether is it spaghetti or course meal), how you deploy it(how many clusters, how many database servers, how your queries get passed) all these matters alot if you want to go big.
So far I agree with what you have conveyed. Indeed there are much deeper consideration going big.

It is best you think of deployment in long term. I give you a very good scenario if you are going hosted and your site gets so awesome that you want to do some data analysis from your users, you decided to deploy a data warehousing system. You begin looking at options and found this bloody cheap and awesome AWS Redshift that costs only $1000/yr/tb (Amazon Redshift) . But when you read the faq, you realized that
in order to import your data you need to either use S3 to do it free or sent it via data pipeline. If you intend to use data pipeline this deal doesn't seem so great anymore as you will need to incur additional expense.
How is it so, just because you need to incur additional expense if you choose to use data pipeline and it becomes unattractive anymore ? Doesn't seems to be what I understood about Redshift. I think if you today compare Redshift alone to any traditional large sized BI solution, it is already a steal. The fact that you can shutdown the instances and then save your cost, how would a traditional data centre solution be able to save on the invested cost ?

When you say using data pipeline, may I understand where is the source of your data transfer ? Are you referring to the on premise data ? If so, would AWS Import be feasible for your scenario ?

Even so, you might want to be more thorough in your TCO. Having your own BI solution besides the software licensing and the infrastructure investment, you also need to invest in system engineers, solution architect to keep the system in tip-top condition. It's not like Redshift doesn't need a solution architect, but the need for system engineers managing a on-premises solution may not be required after all. Even if they are indispensable, the cost in getting personnels with niche skill sets might not be as important using Redshift versus traditional approach.

Every vendor has their pros and cons, I don't support any because I use it based on what I am trying to do. Heroku is good because it is easy to deploy and once you scale out you don't have to waste so much time thinking about how to configure multiple servers. Some people might say its expensive but depending on your case, you might saved it on labor costs.


I am a wheel dealer, what I do is I get a dirt cheap hourly priced hosted or cloud service like AWS for development (why aren't you using that? heck you can do all sorts of funny things).

BUT when it comes to real deployment, I make sure I planned carefully so that I don't incur so much technical debt (Technical debt - Wikipedia, the free encyclopedia) .

Unfortunately if you want to build a successful app, you have to at least understand what is the best way to deploy your site. Okay, back to writing code.
It's definitely good to know how to position your solution not only technically, but also understanding the business concepts involved. After all technology is a business enablement tool. There is no stopping from one to use several cloud solutions in the whole solution ecosystems, as business and solution redundancies. Just calculate the cost and efforts require to sustain such an approach. If it make sense, why not. There are always times to be prudent.

Last edited by davidktw; 17-04-2013 at 01:17 AM..
davidktw is offline   Reply With Quote
Old 17-04-2013, 01:40 AM   #45
Arch-Supremacy Member
 
davidktw's Avatar
 
Join Date: Apr 2010
Posts: 10,044
"No wonder the IT standard in this part of the world is going downhill...." I do have similar observation on Singapore's software developers. I think this is mainly due to lack of interest in software development. Programmers don't get much respect here. And most of the Computer Science graduates are not really passionate about programming in the first place, programming is just a job to many programmers here. I can understand why Software Development field is going downhill. This article pretty sums it up quite well: Rise of the Software Craftsmen - SGE : SGE
I just thought I want to put this as a separate post since it is a relatively different nature from what TS wants.

Now about this software development endeavour, what I feel is people nowadays is greedy. The greed is want more than what they are worth. Reading other forums in HWZ at times, I read about some whom are either still studying or just graduated asking how much they are worth in the industry and how much they should ask for in new company. Then they start comparing whom someone they knew got a higher pay than what is offered to them.

My first thought is "what the heck ? Just a piece of white paper..." If you ask me, 20 years ago when poly students and degree holders are cream of the crop, perhaps such conversations make sense. Today where annually universities are churning out tens of thousands of fresh graduates, perhaps it's time to set a more realistic expectation.

Holding on to the fresh piece of certificate means nothing. It just means you got a entry ticket. To make it in the industry or break it depends on first and foremost ATTITUDE, second COMMITMENT, and third CAPABILITIES. Without the right attitude, you drown before you start. Without commitment, you drown before you are halfway in the swim. Without capabilities, you can't swim far or even reach the land.

But so far, when I read about those comments asking how much they are worth, immediately they failed in ATTITUDE. Fortunately or perhaps unfortunately, not everyone is cut out to be a technologist. If they have no passion, then we don't need them.

Respect is something to be earned. If you want to be respected, then learn to appreciate in what you are doing and respect what you are doing. It's not just programmers or not, it's in every industries.

I think we need to be realistic too. Programmers are high level construction workers. While building construction workers work their asses under the sun, programmers work their asses around the clock. They are different kind of hardships. Face it, the only roles in this industry that is going to earn some prominent respect are managerial roles. If you want to be a big time in this industry, know how to command. A good commander have a lot more to offer and larger influence in the company. When I say managerial roles, it doesn't need to be a manager or a director. It can be a good technical project manager/director, a good solution architect or consultant, a good tech lead, or maybe a CTO or CIO. I know not all candidates of these roles live up to expectation, but if your aim is to excel and have bigger influence over things around you, these are the designations that will give you some level of satisfaction. So it seems in this part of the world, that's how it operates.

Something we really need to learn from the westerns. When the rest of the world compete with them, their attitude is we are better ? (Whether this is true or not is not the point). The point is the ATTITUDE! "We are worth more and we are better because we can deliver better!" That's that kind of impression you get from them.

But here is different. When our rice bowl is shaken... Why are you fighting with me ? Why are you underquoting my worth ? I hate you because you use a more affordable price to fight with me.

See the difference in attitude ? It's not how foreigners manage to get into our industry. Instead of having fighting spirit to hold your head high, to make your stand in the homeland, to have attempts to improve your own worth, it ends up questioning why others can do the same job as you are at halve your price. Instead of telling your boss that you can deliver more, you end up asking your boss why he choose the foreigners over you. That's where it all fails seriously.

When the strong comes, the correct attitude is we must be stronger. Fight back with skill sets, fight back with knowledge. That's how things will improve. Hence with respect to the article found at Rise of the Software Craftsmen - SGE : SGE, my opinion towards foreigners coming into our land and taking over our place, sometimes we need to ask ourselves this question.

What have we done to deserve more ?

Last edited by davidktw; 17-04-2013 at 01:52 AM..
davidktw is offline   Reply With Quote
Reply
Important Forum Advisory Note
This forum is moderated by volunteer moderators who will react only to members' feedback on posts. Moderators are not employees or representatives of HWZ. Forum members and moderators are responsible for their own posts.

Please refer to our Terms of Service for more information.


Thread Tools

Posting Rules

Smilies are On
[IMG] code is On
HTML code is Off
Trackbacks are Off
Pingbacks are Off
Refbacks are On