Elementor #33

Everybody’s doing the same thing. We just do it better.

In today's modern IT and DevOps environment, many companies compete for the same markets. The vast availability of OpenSource or proprietary tools allows complex environments to be brought to life.
Everyone can do it.
It's just that we do it better.

What makes us better?

Experience. It's easy to say, but we actually stand for it.
We have huge experience with small and simple to huge and complex projects for various companies across Israel.
We were there to guide, design implement, and yes - even bash our heads at complex bugs found in various products.
You know that leaflets are great, however, real products look slightly different. We know that too - we have had a lot of experience with that. From Linux bugs, to Oracle GI (ClusterWare/RAC) crashes, through various HA undocumented "special" cases - just name it.

What do we do?

We do not do everything. We specialise in specific technologies, and the whole design which should come around them.
Let's emphasise - design - it means more than just throwing bunch of technologies at a problem. It means taking the time to consider the implications of managing these solutions in the short, mid and long term. It means taking into consideration the ways to implement their knowledge and knowhow into the organisation, it means balancing the use of a well-proven existing solutions, which might provide only "most" of our needs, with writing from scratch a whole new solution which does exactly what we want, and no one in the entire world knows, but us.

Linux installation and troubleshooting

We know our Linux. We know it really well. Although the company is not a development-based company, we know how to handle kernel modules, to some extent ; We know how to identify a problem ; We know how to tackle a performance issue, and how to put our finger on it. Since we have reasonably good knowledge of networking, excellent knowledge of storage and some knowledge of databases (not to mention all the other stuff you will see below), and since we bring experience - we are capable of identifying and solving problems fast.

High Availability

We've been doing HA for years now. We've had experience with IBM HACMP (brrrr), with Veritas Cluster, (even with MSCS) - so we know our stuff. We deal today mainly with RHCS (both versions 4-6 and PCS found in RHEL7), and Oracle Grid Infrastructure (GI). The later as the basis for RAC. We can HA any service, and we've done it for many places.

Storage

We understand storage, so we can build one. No, really. We can connect your systems to existing storage devices, and advise about the best method to connect it (fabrics, networks, multipath, and more - see performance above), and we can also setup a storage for you. We're great fans of ZFS, when the right conditions are met, and we're not afraid of using it. Actually - some of our customers have an in-budget storage based on commodity servers and hard drives, with excellent performance and great availability (see above "High Availability").

Oracle RAC

We are not DBAs. We don't know all the SQL magic, and we do not want to. However, where it comes to RAC, a stable and firm cluster is required. And to setup Oracle RAC Grid Infrastructure, you need to know how to handle the prerequisites, and how to manage the environment around the system. We are the builders of some of the larger and more critical RACs around Israel. We are signed on some of the more loaded ones, and, alas, we can also troubleshoot many of their problems.
It also comes related to storage handling, as Oracle RAC makes use of ASM, and we understand ASM.

UFODB

We need a whole page about it. It's our version of Oracle ODA (and ours exists in the market about a year earlier!). If you know it, you know how nice a box it is. If you're not - this is a fast little box (2U in size, in our case) with about 7TB SSD net capacity for your DB, at amazing redundancy and availability (capable of losing a node, a disk, a power supplier - what not?), with about 300,00 0 IOPs, or so. Who counts when disk stops being a problem in your RAC. This innovative system is serving some of the largest, most loaded systems in Israel now. This whole topic needs a dedicated page. TBD, right? Who has time when you're actually doing stuff?

Automation

Almost everything you're doing manually - you should not be doing manually. Scripts are your friends, and we firmly believe in using them for our purposes. We write mainly in Bash (although we rarely do some Python or Ruby), and there was one time (or a few times, who counts?) when we've produced a script using PowerShell, because, as you can read above - the right solution for a task is not always the one we find the most comfortable.
We automate system installation, system deployments, database deployments (yes, even Oracle), special hardware deployments and so on and on. If you can run it, we can automate it. We love to use commodity tools (on that below), but we will use scripts if the environment requires that, aka, no automation infrastructure already exists, and no intention on adding one. We also do DevOps, but on that below.

DevOps

One of the hottest words today. DevOps. Who doesn't do it?
Well, we do it. We are experienced with Foreman and Puppet, because they go well together, and we've done many deployments of this combination (including upgrades, after careful planning, of course), and some deployment automation with Jenkins. Of course we've done some magic with AWS (we don't do Azure, sorry, not yet), and we've managed to overcome and workedaround some of its more troubling limitations (try writing fast to S3). So we know this stuff too. There are tons of buzzwords, but we try to pick our own carefully. Many appear for a day or two, and then fade into the black. Also - we cannot keep up all Amazon's new service names here, because we will be updating the site all the time, instead of actually doing some work.

Service

Now, this one is important. While it's not a technology thing per-se, this is an important thing. We provide service and availability for our customers, with 24x7x4 response (and in some cases, upon request and reasonable compensation - even better), and we deliver. We try to earn that money by being available, but doing so well and delivering the systems as they were meant to be (TM), we sleep well at night. However - we can be called and we troubleshoot, even in the middle of the night. We are aware of your business needs, and we are aware of the implications of downtime. We do not want you to have one, and if you do, for some reason, we will help make it the shortest possible downtime.

Our principles

We do the best work possible. We cannot do it any other way. It means that if you use us, you get the best possible work. Period. We also work fast, and efficiently, and we will not do things we do not believe in. So the whole principle of "the customer is always right" doesn't always apply - if the customer is wrong, we will tell that to the customer, and, politely, explain why. If the desired solution does not meet up with our standards, we will not do it.
However, if it does - we will gladly do it. And with a smile. We like to take things lightly, and make projects fun, while still delivering on time, exceeding expectations. Don't hesitate to expect. We will soar above your expectations, and make you say that cool "Wow"

These are some of our clients

Barnoy Clients