Amazon Redshift Reviews

We stored all of the data in the S3 bucket and would like to have it stored in a data warehouse, which is why we chose this database. 

It would be very easy for us as an end-user, who would like to access the data, rather than draw it post-transformation and store it at a database level.

The TP transactions for the creation of the tables does very well.

It is quite simple to use and there are no issues with creating the tables.

The managing updates, deletes, and role-level change performance is very low. For example, while you are doing inserts, updates, deletes, and amalgamates, the performance is very, very poor.

If you want to query the database after you have a lot of terabytes of data, the load, performance-wise, is very low.

Looking at the performance of the query, querying the database, and especially with the amalgamates when it is getting updated, it is really poor.

We like this solution and have tried all of the native services; they were working quite well. The only concern about Redshift was managing the cluster, especially the EMR cluster. Our company policy was not to use EMR clusters, especially with the nodes failing. There were many instances of downtime happening. Essentially, there was too much data traffic.

The other drawback was the CDC, as we do not have any tools that can support it.

Creating the structure is easy on the DDL side, but after you create the table and you want to transform the data to store it in a database, the performance is poor.

It takes a lot of time to ingest and update the data. After you ingest the data and someone wants to fetch it in the table, it takes a lot of time performance-wise to return the results.

We have been using this solution for three months.

We are using the latest version.

There are issues with stability and it should be compared with Snowflake.

This solution is scalable. We scale up and scale down manually when we are required to, we do not have an automatic setup.

We have three or four people using this solution.

We have contacted technical support to give our opinion and recommendations or feedback and they agreed that it needs improvement.

Previously, we tried the Snowflake database, which works really well. The expectations were really good with the performance, also the DDL, DML operations on the processing of the data.

The initial setup is simple and we did not find it very complex at all.

The time it takes to deploy depends on how many tables you want to create, or how many tables will you merge the data with.

We are switching to Azure, although not because of the product or the services that we did not like. It's about AWS being competitors for logistic companies that we are working with. Also for security reasons, we do not know how secure the data is on the cloud.

If you are competitors then you don't know if the data can be accessed by your competitor, and the team can be looking at a demographic, which could impact your sales.

We have only just started using Redshift, but we are not really satisfied with it.

I would rate this solution a six out of ten.

If you want to create an enterprise data hub, that is where Redshift is used. Snowflake, Redshift, BigQuery, and Azure Synapse are enterprise data warehousing and cloud data technologies. Large enterprises have enterprise data. They have a lot of managed processes, business processes, customers, products, different assets, locations, equipment, etc. Then they have sales and marketing. There's a huge amount of data that is generated, and they will need a large warehouse or multiple data warehouses to create analytics out of that data.

We try to tell organizations to consolidate all their data into a single unified data platform that has all the enterprise data rather than being processed by multiple warehouses. It's processed on one central data platform, which is cloud-based. In which case, we recommend one of these four. We either recommend Snowflake, Azure Synapse, AWS Redshift, or Google BigQuery. It depends on what their early investment is and what kind of work they need to do.

Redshift is completely Managed on AWS cloud.

I would like to see improvement in the pricing and the simplicity of using this solution.

The product is very stable, and so are all other cloud-based managed Enterprise data platforms (Snowflake, BigQuery and Azure Synapse)

It's scalable as it's on hosted and Managed on AWS cloud.

Technical support is great, very professional.

Neutral

I would recommend Snowflake the highest, then Google BigQuery, Azure Synapse, and then Redshift.

If somebody is heavily invested into Microsoft, then going for Azure Synapse is what we recommend. If they're open to moving to a completely new system, we evaluate the landscape and we recommend either Snowflake or Google BigQuery. What we recommend and what we design and create and implement for our different enterprise customers is very different for each customer. There's no One-size-fits-all solution.

For example, for one of our customers, we have helped design and create their entire single unified data platform using Snowflake.

I would say Redshift needs a little more effort and expertise for setting up the kind of infrastructure one need. If you can do something with two-three people for Snowflake, you would need four people on Redshift. You need to have a little bit of knowledge of the AWS Cloud and AWS services to be able to use Redshift. A typical Redshift based Enterprise data work would need anywhere between 4 to 15 people.

The return on investment of moving from an on-premise to a completely cloud-hosted data platform is significantly high and worth the effort.

Redshift is costly compared to other solutions.

It's pay per use. You can have multiple models. You can go for yearly cost, which is a little discounted than the monthly cost. Depending on how much data you process and store, you can have different pricing. There's no fixed cost. All of these are based on how much data you store monthly and how much data you process.

I would rate this solution 6 out of 10. 

If an organization has invested heavily in AWS services and they have a good knowledge of the AWS ecosystem, then I would recommend Redshift. Otherwise, I would still recommend Snowflake because Snowflake works very well with AWS services. I can have my AWS S3 buckets in which I can store my enterprise data lake, and then Snowflake works with that seamlessly. If the organization has good knowledge of AWS and good knowledge of RDBMS data warehouses, then we can recommend Redshift to them.

It all depends on how much investment that organization has done in Redshift. For example, we have a customer which has a very large setup. It's a large US-based company, where they have invested heavily in AWS. They're an AWS house, so they like everything about AWS. For them, we have recommended Redshift so that the overall tech ecosystem remains optimum. 

For the on-premises version of Amazon Redshift, we need to start from scratch. However, with the cloud version whenever you want to deploy, you can scale up, and down, and it has a data warehousing capability. Redshift has many features.

They have enriched and elaborate documentation that is helpful.

We are using third-party tools to integrate Amazon Redshift, they should create their own interface on their own for it to be easily connected on the AWS itself.

I have been using Amazon Redshift for one year.

Amazon Redshift is reliable and has high availability.

The scalability of Amazon Redshift is good.

We have approximately 20 people using this solution in my organization.

Before we were using Amazon Redshift, we were working with Postgres, Greenplum, and Oracle SQL. These were on-premises databases, and we migrated to the cloud.

The price of Amazon Redshift is reasonable because it depends on the usage that you use and for DWH for the long term.

I would advise others that if they have a large set of data where you have a less number of updates, then choose Amazon Redshift. If there is more update and fewer inserts, then do not use Amazon Redshift.

I rate Amazon Redshift an eight out of ten.

We use the solution to build a data warehouse schema for a target database for analytics. We are uploading data from different transactional databases into Amazon Redshift. We use it for reporting purposes. We use the tool mainly for querying and retrieving the data for analytics.

The fast querying of a huge amount of data greatly impacts our data workflows. All the queries work pretty fast.

Amazon Redshift is a really powerful database system for reporting and data warehousing. I like the product. It works really fast with significant volumes of data. The product covers all the main functionalities required for our data security and compliance needs. It has almost everything we need. It is the main data source for our analytics functionality. We can run our models using the data stored in the database. The ease of use is fine. It is pretty easy to integrate the solution with other products and third-party solutions.

The product must provide new indexes that support special data structures or data types like TEXT.

I have no complaints about the product’s stability.

The tool is scalable. About 30 to 50 analysts use the solution in our organization. We need one or two people to administer the solution.

I haven't heard any complaints about the support team from our DevOps engineers.

My project involves analytics and data warehousing. I use Amazon Redshift. I also use AWS Glue as an ETL tool.

I will recommend the product to others for data warehousing and data analytics. However, I do not recommend the solution for small companies that do not have enough volume of data to analyze. Overall, I rate the product an eight out of ten.

We use this solution for production and customer data for one of the lines of the business, which is basically one of the warehouses. I oversee the whole architecture infrastructure.

The price isn't bad for the performance for a cloud data warehouse. It's also connected to Databricks but uses SQL. It's comparable to BigQuery. If the analyst knows SQL, which is comfortable and easy to use to go between all of these tool stacks, I think it's reliable. It's a secure and reliable data warehouse.

The performance is good, and it's pretty fast. We also have Looker and MOLE connected to it for visualization, which works seamlessly. We're storing a lot of data. There's a lot of transactional data, clickstream data, and telemetry about the customer, what they're purchasing, call logs, and marketing data.

The analysts are familiar with SQL, and they're able to do this. Even the data scientists who aren't that savvy in Python, because they are very strong in SQL, are able to interact with it very quickly. I'm able to bring in more analysts for support.

There are physically too many pipelines for a company of this size to maintain. For a data scientist, it's very difficult to learn the data in all of these different environments. It's easier to train people in just one environment to start with, like Snowflake or Databricks. It's difficult to have so many technologies that are very comparable, and each comes with a price tag.

I have used this solution for seven months in my current company, but I have also used it in a couple of my previous companies.

The stability is good. For a mid-size company, it's very stable.

It's very scalable.

I would rate the technical support as 10 out of 10.

We had one or two tickets, and they responded extremely quickly. Technical support is good, but that's because we aren't running into many issues. The solution is pretty stable. The individuals who set it up did a very good job.

It wasn't too complex because only a few people were needed to set up the solution.

It's not very pricey compared to other tools. I would rate the price as 5 out of 10.

In my organization, we also use GCP, Amazon EC2, Databricks, and Snowflake. We also have Delta Lakehouse. I'm going to move everything into the Delta Lakehouse. For a company of this size, it's a lot of tools to physically maintain with a small data engineering team.

I think Snowflake has a few more features. In Redshift, you need to write a bit more SQL in some instances, but it's very user-friendly and fast. It can be used as a data warehouse solution as well. It can also do some analytics.

Redshift is comparable to other solutions. I wanted to go with Amazon EC2 because we also have Databricks, and I think I can cover some of those features with the combination of that.

I would rate this solution as eight out of ten. It's a very strong solution.

My advice is to do your research and see if it makes sense for you. You can always request a demo from Redshift.

We mainly use the solution for marketing analytics.

The solution works fast. I use Redshift to clear a lot of web page data. I use it mainly as an extraction tool to obtain the information I need for a project and store it in parquet files on a disk. Later, I work on the data using Python. I write back all my final results to Redshift and store the temporary files on a local machine.

They should provide a structured way to work with interim data than to store it in parquet files locally. Also, Redshift is unwieldy. There should be better integration between Python and Redshift. It could be more accessible without using so many sequels.

They should make writing and reading the data frames into and from Redshift easier. The performance could be better. I have used Redshift for extensive queries. For the large tables, it's easier to unload to Redshift, but subquery tables that run complex grids are slower for configuration. I have to use the unloaded command to unload the whole table. Further, I have to read the table into a server with extensive memory in Python and process the data ahead. It's not optimal. 

We have been using the solution for two years.

I rate the solution’s stability as a nine.

The solution is scalable. It handles different loads very well. We have 80-100 users using the solution in our company.

The setup was quite complicated. For instance, if you use AWS Glue to automate loads into Redshift, setting up the security for the requirements between the two is complex. I've struggled a lot to set up the cluster on VPC and to get all the endpoints set up correctly with the right access and services. Especially from Glue's endpoints, I had to repeat the same process every time. It consumes a lot of time. In comparison, the CloudOps executives do the setup very quickly.

I have heard complaints about the solution’s pricing, and thus I rate it as a five.

I rate the solution as an eight out of ten.

We were using the solution for our data backup, but we wanted to optimize it, so we turned to AWS Glue. Amazon Redshift wasn't really great for us and wasn't working out.

Amazon Redshift was used for data storage while moving back from S3 to Amazon. However, it lagged, taking its own sweet time for data backups which also depended on the server location. Because of the aforementioned reasons, we started losing a lot of data that wasn't even real-time data. Ultimately, this affected our analytics at the end of the day. Also, we have been trying to do some work on our AI models, which emit out recommendations based on the live dataset. There were a lot of lagging issues. So, for example, sending out somewhere around 0.1 million or 100,000 emails used to take almost 12-14 working days, and this also includes the process of pulling all the data and sending them to CronJobs. Since we wanted all this work to happen in real-time, we had to get rid of the tool.

I would like Amazon Redshift to improve its performance, analytics, scalability, and stability. Other than these points, I am not aware of any other areas to address since Amazon provides a variety of independent services for their customers to choose from, and if one were to express dissatisfaction with Amazon Redshift, Amazon would likely suggest AWS Glue as an alternative. Similarly, if another issue arose, Amazon might recommend Amazon RDS. There are a lot of things they try to upsell to you, each with its own pros and cons and in different packages offering different perks. So, it all depends on your business needs and what you choose for your business. I wouldn't criticize Amazon for this because they have created packages tailored to their customer's needs, which helps to prevent customers from looking elsewhere.

I have been working with Amazon Redshift as an implementer for three years.

Stability-wise, it has a lot of issues with threats, and that is why we went for a threat shift optimization. In short, I mean to say that it is not stable at all.

If you require a highly scalable solution, I would not recommend Amazon Redshift. We currently have 12 clients using Amazon Redshift, and the scalability of the solution is terrible. In terms of scalability, I would rate this solution a three or four out of ten.

Amazon Redshift has a lot of documentation, but the setup is fine. Three years ago, the solution's deployment process took over a month or a month and a half.

Every solution has a cost and comes in different packages. Considering these factors, AWS Glue is on top. Though Amazon Redshift is good, it depends on what kind of business you're trying to do, what type of analytics you need, and how much data you have.

For Amazon Redshift, we pay around INR 60,000 annually. The cost factor also depends on the number of existing customers. In addition to the standard licensing fee paid for AWS, we incur a cloud storage cost of around a quarter million for the amount of data. We also have to bear additional costs for data security and cybersecurity, which are well taken care of by Amazon, hence the premium pricing. There are several other features and services provided by Amazon that justify the premium pricing.

Amazon Redshift is a horrible solution. I recommend my customer to use AWS Glue since while dealing globally with real-time data, which you need to make decisions, factors like how much cost and data is needed to make a decision should be considered. Apart from this, if customers are paying a huge price for the solution, then probably Amazon shouldn't mind spending on the tool. However, it may not be necessary for small businesses with only a few thousand data points. Although Azure is a better option, some clients prefer AWS, and we had to develop a solution using AWS for our client. Overall, I rate this solution a three or four out of ten.

It is scalable, easy to use, and has very good query performance. It is serverless as well.

They introduced machine learning directly in to Redshift, and you can now query using machine-learning functions.

It's very easy to migrate from other databases to Redshift. There are migration tools dedicated for this purpose, enabling migration from other databases like MS SQL directly to Redshift. The majority of the scripts will be automatically transposed.

AWS Snowflake has a very good feature for cloning databases. It makes it easy to clone a data warehouse, which is useful. I would like to see this feature in Redshift. 

We have been using Amazon Redshift from 2013.

For stability, I would give a rating of ten out of ten.

We have 16 Redshift users in our company, and I'd rate the scalability at ten out of ten. 

Redshift's technical support has been great.

The initial setup is simple.

In comparison to the price of similar solutions, Redshift's cost is the lowest.

I would recommend Amazon Redshift and would rate it at nine on a scale from one to ten, where one is the worst and ten is the best.