Closed. This question is off-topic. It is not currently accepting answers.
Want to improve this question? Update the question so it's on-topic for Stack Overflow.
Closed 10 years ago.
Improve this question
Does anyone know of inexpensive "origin pull" CDN service providers.
The only provider that I've found that provides this are SimpleCDN and Akamai. Akamai is crazy expensive and SimpleCDN seems to change their business model daily, so I'm concerned with using them.
We use PantherExpress in "pull mode" (not sure they do anything else, really) and it's great. Also very competitively priced. They're now owned by CDNetworks; I haven't talked to their sales people at all so I don't know how much changed.
LimeLight were also much better priced than I expected when we were getting quotes; but they had this stupid "oh, a little more for this feature and extra for that feature" pricing where PE just gave me a simple price including all their features.
When we were evaluating EdgeCast was missing some of the features we needed, but I think they've caught up now.
We use 10-20Mbit/sec on the CDN to give you a ballpark.
Few other CDNs:
Edgecast
Highwind
Reflected Network
Afaik most of pay-as-go CDN providers, as CDN77.com or Maxcdn, has this feature as the simpliest and standard one. You can find more about Origin Pull providers also in discussion CDN: "Origin pull" service providers?.
As I'm using CDN77 I can confirm you that they are providing this without any problem, with 14 day free trial version, on about 50 or 60 PoPs.
I've used LimeLight with an origin pull model and it works quite well. The only issue is their edge network is constantly expanding so if you want to employ any kind of security for your content on the origin (i.e. firewall ACL's) it becomes a constant maintenance PITA. OTOH, if your content if wide open to the public then I would highly recommend them.
I don't know how cheap they are for a new customer. I guess it depends on how much traffic you are expecting (and ultimately how much money you will give them). My company does a fair amount of CDN business so we got really nice pricing, but YMMV.
Also, if you are a Rackspace customer you might be able to leverage their pricing - they use LL exclusively for their own content and for their cloud offerings (http://www.rackspacecloud.com/cloud_hosting_products/files).
Good Luck!
You might want to give VPS.net CDN a try:
https://www.vps.net/cdn-signup
According to their wiki they support "Origin Pull".
I don't work for them
I haven't used their service
but their pricing looks very competitive.
Related
Closed. This question does not meet Stack Overflow guidelines. It is not currently accepting answers.
This question does not appear to be about a specific programming problem, a software algorithm, or software tools primarily used by programmers. If you believe the question would be on-topic on another Stack Exchange site, you can leave a comment to explain where the question may be able to be answered.
Closed 8 years ago.
Improve this question
We're making a lot of websites for multiple clients and we always had some problems figuring out what to do with "Updates". We're starting building Wordpress sites and there's almost a plugin update every week. Since our sites use WPML and Woocommerce, there's a lot of conflicts happening with unreliable WPML updates and other plugin with security flaws (revolution slider in my case).
I've just received an email from my client's hosting and he wants us to make updates from WP 3.8.1 to 4.0 but I know there's an intensive conflict job coming up. I usually use the 'Disable all wordpress updates' plugin to hide 'em all from the wp-admin.
I wish my project had enough budget to build custom sites, but that's not the case here.
I just want to know how Wordpress agency are with that. It doesn't look that professional to tell a client that updates aren't necessary.
First of all, you of course bill them for the time you spend on your clients websites. Better yet, sign some kind of support support agreement where they pay you a monthly fee for having their websites kept up to date and maybe uptime monitoring. You can also include a few hours per customer for support and small development tasks per every month(it's usually nice for the clients to have even if they don't use it every month). This is a nice way to get some steady income and fill the time gaps when you have less to do.
You should also state in the agreement when you build the website that future updates and fixes is NOT included in price. You can give the clients a short period the report errors and bugs in the website and after that you will bill them for any extra time.
Secondly, get the right tools to optimize your time. I recommend ManageWP or InfiniteWP. ManageWP is slight better in my opinion and it is a hosted solution, but has a higher price if you have many clients(per client pricing model). InfiniteWP is a free self hosted solution(you set it up on your own server), where you only have to pay for the modules you use. You will need a few modules to get the functionality you need, but it's still much cheaper than ManageWP if you have many clients.
For uptime monitoring I recommend Uptime robot(simple and free for up to 30 websites) or Pingdom a better service with a lot of nice tools, for example performance monitoring, but also a much steeper price.
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 8 years ago.
Improve this question
I have a client whose requirement is best met with an XQuery/XML solution. The problem I am facing is overcoming the risk associated with a lack of market place skills for these technologies.
This is maybe a sales question, but how have others overcome this objection?
I'll let someone else answer from a sales perspective or suggest technologies. Here's my project management perspective. I think you should do two things:
1.) Cost of ownership assessment
Draw out two or three architectures and try to amortize in hours, $$$, or some other quasi-imaginary metric the immediate and ongoing impact to the client. For each solution, how hard will it be to build? How many different engineers will you need? How many different skilled people will you need to keep familiar with the project to maintain it, etc. Does the benefit of not having to have separate middle tier and dedicated relational database people outweigh the market availability of XQuery people? You have identified that the problem is best met by XQuery/XML. Can you quantify this somehow to your client?
2.) Risk mitigation brainstorming
The idea here is to come up with a plan to reduce the possible impact to the client for the technology that you choose:
Start with a Proof of Technology project to gauge the difficulty and efficiency benefits for your staff / client to implement an XQuery solution
Develop an in-house expert / team:
XQuery Training is available
Share cost of supporting XQuery expertise with other projects and similar problem spaces at your organization / client
Expose the XQuery/XML portion of the solution through other means that don't require special skills
XML and/or JSON over REST
Some sort of data access object layer that doesn't remove the agile benefits of the XQuery technology
SOAP services (yuck)
Build a relationship with a service provider who knows their XQuery well
From more of a sales perspective it is important to realize that there is actually lots of XML expertise in the market. XQuery expertise, while perhaps more widely available than you might think, is more limited.
I would look to find an XQuery/XML technology that meets a couple of important criteria:
Numerous successful customers who are willing to act as references. Pointing to successful implementations in production and, even better, allowing your customer to talk with some of these implementors, can be very powerful in reducing the perception of risk.
Provides development tools that allow for rapid development and reduce the amount of initial XQuery knowledge provided. For example, look at this open source framework where you can access nearly all of the underlying XQuery functionality via REST services or even via language specific APIs built on top of REST: https://github.com/marklogic/Corona
Have a robust training program that is accessible and reasonably priced. The cost of training is often low and the risk reduction is high.
Have an active community of developers and open source projects. This will allow you to leverage a lot of existing work done by the community as well as general development knowledge resident in the community, further reducing risk.
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 6 years ago.
Improve this question
In our organisation we deliver products to different product lines depending on the requirements. in short the same application is customised according to customer requirements and delivered. After deploying the application sometimes we got some issues logged by client.
My question comes here. who is responsible to look into the issues and solve it
Programmers
Testers
Management is asking Testers to have a look into the issues and solve them. But the testers don't have the chance to look into the code. is it feasible to ask the testers to go for the issue resolution and end up wasting time doing nothing thus delaying the solution to the customers.
I would normally expect management to look through the issues every so often (say, every week), and allocate depending on schedules, severity, forthcoming releases etc. Some questions are:
is it an issue a bug, a feature request etc.?
does it prevent your client from working with your tool ?
is it impacted by forthcoming work (e.g. will a new feature remove the feature causing the issue) ?
I don't believe you can resolve these issues in isolation. It requires project managers etc. with awareness of project direction and programmers with awareness of the codebase to work together to determine how/when issues should be addressed, and their impact on other work streams.
Initially you should have a support department that does triage on all newly added issues. They should be empowered and informed enough to decide whether this is a non-issue, whether there's a work-around or whether they don't know. If it's the latter then it should be elevated to programmers.
You might also want to include the testers in the chain if the support guys are unable to produce an adequate 'how to reproduce the problem' document for the programmers.
The way it works at our company is that the testers are asked to verify the client's issue, i.e. trying to reproduce it and document the steps taken to reproduce it. Then it gets logged as an official bug and assigned to a developer who can retake the tester's steps and hopefully fix the bug.
Testers can identify an issue. How can they resolve the same? Only the developer will be able to do it. Looks really strange where a tester is asked to resolve the issue.
Who deals with the clients? Liasing with clients is not a task normally associated with the technical staff.
You should have someone whose role it is to speak to the customers, find out exactly what the issue is and how the client would like it resolved so that it may be passed onto the most relevant person to address the issue.
I would say the logical way to do it is:
Testers should try to reproduce the problem and identify its source
Report the problem with steps to reproduce it to the programmers
It's not common usage to let testers solve the issues as the programmers won't get the feedback they need to avoid the issues in the future.
Testers - verify that the problem exists.
Programmers - solve the problem.
In between there is another part to this, which is "gather information about the problem". Usually this is a split between testers and programmers; exactly how balanced that load is depends on the team.
If you don't have the code, you can't fix bugs. It's as simple as that. At the very most you could fix configuration errors, but if the misconfiguration was caused by the program that's a short-term fix.
Closed. This question is off-topic. It is not currently accepting answers.
Want to improve this question? Update the question so it's on-topic for Stack Overflow.
Closed 10 years ago.
Improve this question
I know there are a million ASP.NET hosting options, but what are the premium options if you have some money to spend and want maximum performance and uptime? We currently use MaximumASP and they are generally great. I know another good option is Rackspace. Does anyone have any other suggestions? This is one of those things that is hard to Google, because everyone calls their hosting option premium or professional.
I'd say Rackspace is a good choice, but I use discountasp.net because I needed .NET 3.5 SP1 hosting with SQL 2008 and they delivered.
If you want something between shared hosting and Rackspace dedicated, check out their newly acquired cloud offering. I've been using it for a few months with a lot of success.
http://www.rackspacecloud.com/
The obvious choice would be to get a dedicated server, other than that I love: DiscountAsp.net as they always have the most up-to-date frameworks on their servers. Useful if your trying a new technology such as when MVC was in beta
I have used SoftSysHosting for a multitude of clients, and they have never let me down...plus, their Customer Support is excellent, accompanied with a Knowledge Base of FAQs.
I use ORCS Web for one of my sites. I've never had any problems with them in over 2 years. They asked me to fill in a survey a few months back asking what I thought of their support and I had to admit that I'd never really had to use them beyond the initial setup.
You should consider AppHarbor. AppHarbor provides sophisticated scaling and load balancing and a catalog of great add-ons. We also provide tools to make it easier for developers to move their code onto the platform using their favourite revision control tool (Git is supported and the platform is integrated with Github, Bitbucket and Codeplex). Here's a good overview of How AppHarbor Works.
(Disclaimer, I'm co-founder of AppHarbor)
Terremark.com
http://www.terremark.com
Managed Hosting and Enterprise Cloud options.
Closed. This question is off-topic. It is not currently accepting answers.
Want to improve this question? Update the question so it's on-topic for Stack Overflow.
Closed 12 years ago.
Improve this question
Recently a friend of mine had gone from a high level NOC position to a developer. Before that he was just doing the help desk stuff. He has no degree, only the usual MIS/networking certifications and as far as I know only tinkers with code on the weekends. I can see where in some scenarios having a good understanding of configurations, packets, users, OU's, etc would be extremely beneficial to a developer.
My question is this, how many full time developers started off this way? Even how many people dual wield the responsibility of developer/systems administrator/network administration?
I'm sure that this is a fairly common scenario. I've spent 12 years in I.T. and I find that as time goes on, the real income comes from being a specialist (DBA, coder, etc.) as opposed to a generalist (network admin, helpdesk).
It's actually the path that my career is taking. I'm not quite a full-time DBA or developer but that's where I'm heading.
I'm also willing to bet that the people skills I've picked up along the way (helpdesk support, network admin, systems analyst) will help me in my DBA/Developer career. Skills I don't feel I would have gotten had I jumped right in to a coding career.
Indeed. I think developers should know the platform they are building software for. If a dev has worked as sysadmin before, he will know how to integreate his software well. Some Windows-Desktop-App related "integration smells" that come to my mind:
App does not run unter normal-user privileges (run on properly secured enterprise desktops? oops!)
App requires write permissions to all kind of system folders (security? oops!)
App stores user settings in 'nonstandard' locations like %programfiles% (backup? permissions? oops!)
App does not provide silent-installable setups (deployment? oops!)
Etc..
A real sysadmin would never write software that has one of the above integration smells. Really.
It's quite common in small companies. I did that for some time - developing the software we sold to customers, keeping the network going, and adding features to the database as needed for a manufacturing company of fewer than 20 people.
You wear many hats in a small business.
But I started off programming microcontrollers in high school, so I can't claim this is where I started.
It is very helpful to have a working knowledge of all these systems as a developer.
-Adam
The overlap of developers and admins happens quite a bit. Our last admin developed on the side just so he'd have a better understanding of what he was helping support. When he left I became the admin just because I tinkered with admin stuff on the side to know how my software was being supported.
A broad understanding with a few focuses is what I'd say is best for any technical professional. Then with a bit of study you can change to meet whatever need may arise.
I've seen it more the other way where a programmer also "admins" the servers and sometimes network. I've definitely been in that position.
I would think it can easily go the other way as well where an admin can start programming systems, but from my experience it's not as common. Whenever I ask a server admin or network person "do you program too?" most of the time the answer is "no".
I think it might be easier for programmers to cross the line because when you are programming a system unless you always have an admin available you need to be able to set up your own environment and that usually includes setting up a server.
I started off as a NOC operator, eventually working my way up to a senior network engineer position. During the last 2-3 years of my tenure at my previous company, I picked up a fondness for programming and started teaching myself everything I could on my own time. Around 2005, I left said company for a small startup and still work there today as as the admin and primary developer.
The one challenge I impose upon myself is to not make admin changes at the drop of a hat to satisfy programming challenges. I must force myself to code in a way that any application I make can be redeployed elsewhere with minimal privileges, despite the fact that I can do pretty much anything I want with our own servers. It's a fine line between performing both duties well and performing one duty badly due to the needs of the other.
I'm here.
Although I've been tinkering with code since I was a child, my first full-time job was being a system administrator, a DBA and other related roles.
Afterwards I worked full time job as a developer, and now I'm both a developer and a security researcher.
Also, I managed to complete M.Sc in CS.
I believe that such transitions are possible, and very beneficial, as you get a wider view on your field of work.