The initial setup is complex, mainly due to all the services that it encapsulates. The complexity depends on the number of components heavily used by a given client or customer. We try to restrict the number of components as much as possible. Deployment took three days with preparation and the downtime was eight hours, which can be significant if you're running a 24/7 operation. We carried out an in-place upgrade. We upgraded the existing cluster from CBH to CDP. It went fairly smoothly despite some challenges. Four administrators carried out the deployment; they are also responsible for Cloudera support.

The ease or difficulty in setting up the product depends on the environment of the customer where the tool is deployed. If a banking, industrial, or retail sector firm is taken into concentration, depending on how big of a database is maintained, including the applications that are to be hosted, the deployment process can range from a simple to a very complex phase, depending on the architecture.

For Cloudera Distribution for Hadoop, one has to go through the usual deployment process, like for any software product. You have to have different environments before going into production, like pre-production environments, test and dev environments. You install and configure all the components in the test environment and then test them on the pre-production environment. Once UAT is done, you move them to the production environment. In general, it's a critical product deployed in a company.

To deploy Cloudera Distribution for Hadoop properly, it may take a couple of months on average. However, for a complex deployment, it can take up to a year. In our case, we had over 12 data nodes and over 30 different servers involved in the implementation.

For the deployment, we require at least five knowledgeable people due to the complexity of the system.

The initial setup is easy.

We had a vendor. They have some issues with the installation, especially with the upgrade. When we upgraded from CDH to CDP on-premise, they had problems with the user interface and user authorization.

Deployment depends on the projects. It takes maybe a month for migration to CDP.

The initial setup was neither straightforward nor overly complex. It took us a few days to complete. I wasn't involved in the configuration; another team handled it. There may have been technical issues related to the clusters. We have a technical team with seven administrators.

I rate the ease of setup a seven out of ten. The deployment takes 48 hours. We need six Hadoop administrators for the deployment.

The initial setup has become easier although you need a dedicated admin to maintain and manage the solution because it's a framework and not a single product. Deployment nowadays is much smoother with the PaaS offering in the public cloud, so you can carry out the deployment with an in-house team. The deployment only takes a day but a company is unlikely to go with the default so the solution needs fine-tuning which can take a couple of weeks. 

The initial setup was complex.

It's not as simple as Oracle Sybase.

It's a complex architecture because you have raw data and many engines.

The initial setup was difficult and we didn't like it. That is only because we implemented it with other software solutions outside Cloudera and needed to do the integrations. 

We are still battling with working out problems with some integrations after eight weeks. It's up and running, but we're optimizing, so that is why I'm saying it's probably medium to complex. But that was the situation for us and our particular needs. It may not be as complex for other businesses at all.

The initial setup for Cloudera Distribution for Hadoop was easy for us because we outsourced the work to the vendor. All the nitty-gritty was taken care of by them.

The on-cloud version is easy to set up. Although, it is complicated to process a large amount of data for on-premises or hybrid setup. It is not a ready-to-use solution for telecom or finance technology. It requires the deployment of robust technology relying on network infrastructure.

The initial setup is straightforward. 

The initial setup was straightforward and not an issue.

The installation is straightforward. We use command-line-based installation and we have created our own way of installing with our product. 

Depending on the customer or depending on internal usage, our DevOps engineer will install it or my development team will install it. 

It was pretty easy to install the product. It took us 20 minutes.

The initial setup was simple, but we had trouble implementing the cables in the Hadoop solution.

The installing is straightforward.

Our clients provide us with the access to use it directly.

Once you have been given access to the edge nodes we are able to run the scripts in the Hadoop layer.

The initial implementation was straightforward from an application side. There weren't any hiccups. In terms of deployment time, it's going to be difficult to say, because most of it was related to hardware problems. Software took about two months to deploy. We required four people for deployment.

Very straight forward. Typical Windows type installation...Next, next, next clicks.

We have struggled a bit in installing and configuring Cloudera Manager on the AWS cluster. For now, it is good.

In the cloud environment where we deployed (Azure Resource Manager) there was a ready-to-deploy template which simplified a lot the initial set-up.

We were already running one production cluster with approximately 75 nodes when I joined, so I’m not familiar with what was needed to get the initial production cluster up. Once I joined, I assisted in standing up the additional nodes and clusters using our chef automation.

Following a single path for installation was initially confusing due to multiple recommended approaches e.g. parcels vs. packages. However, after a while, we managed to master it. However, knoweldge of Cloudera Manager and Hadoop architecture is a must.

The initial setup of Cloudera is difficult. After you have installed it once, it is not difficult to reproduce.

It was extremely easy, and allowed less experienced personnel to get into the context pretty fast. Any difficulties/complexities faced were not related to the product itself rather than to the cluster infrastructure used.

Straightforward. The CDH VirtualBox with preconfigured environment helps for demonstration purposes

It's been quite easy to install. We only had to follow the instructions and there weren't many problems. That's important for us.

It depends on mode of installation. Cloudera Manager is always more straight forward and manageable. Avoid RPM installation as much as possible. Lay out plans with system admin on upgrade plan, commission and decommission nodes. Investigate impact and consequences of having HBase and Hadoop in the same cluster or as separate cluster, what are the impacts on system admin, cost, upgrades, data migrations, resources, etc?

The complexity kicks in when performing parameter configurations. Find out what are the use cases, are there disk IO or compution IO bound, are there lots of structured data or unstructured data for text analytics, etc.

I was not directly involved in deployment.

The implementation of Cloudera Distribution for Hadoop is not easy. It works on multiple nodes and can be complex for testing. The whole process took us one and a half days.

Cloudera Manager greatly simplifies initial setup.

The initial setup is straightforward.

The initial setup is complicated. We needed the vendor to install it themselves. The deployment took around three weeks. Three people were involved because they just follow up and supervise the deployment, but they're not deploying anything. The vendor does it. 

It was complex because we were doing first time deployment of Cloudera on Azure. Also complexity was high due to lot of security features.

The initial setup was complex, due to the user interface. We were doing a POC, so we're still doing the deployment.

It was very easy.

It was very straightforward.