Microsoft Azure Synapse Analytics Reviews

We are using the solution to process a large volume of data. We are taking advantage of the MTPP functionality feature in Azure Synapse Analytics and connecting our reporting dashboard to its dedicated support tool. These are the two primary use cases we are currently leveraging.

Generally, people have differing opinions on whether to use Azure Synapse Analytics or Azure Databricks. These are two different services from Microsoft, and customers often struggle to decide which is best suited for their particular use case. Both services have some features in common and some that are exclusive to one or the other. It is a trade-off and ultimately depends on the use case that best serves the organization. Incorporating the features of both Azure Synapse Analytics and Azure Databricks would meet most use-case scenarios.

The cost of the solution has room for improvement.

Azure Database has a very good Unity Catalog functionality, which is a form of governance within Azure Databricks. Similarly, for Azure Analytics, having a governance product would be great. However, Microsoft already has Azure Purview, which is meant for governance, but it would be nice to have that functionality built into Azure Synapse Analytics.

I have been using the solution for seven years.

The solution is stable.

The solution is scalable. Generally, the development team of around 20 people uses Azure Synapse Analytics directly. Otherwise, a Power BI dashboard accesses the Azure Synapse Analytics database, and the end user uses the Power BI reports, not being directly exposed to Azure Synapse Analytics. Only the development and software development teams use Azure Synapse Analytics.

We are always in contact with the Microsoft team. We keep up to date with our Microsoft reference architecture and other related matters through our communication with the Microsoft team members.

Positive

The initial setup is straightforward. We can have the dedicated SQL up and running within 15 minutes.

The implementation is completed in-house.

The cost of the solution depends on the type of license we choose, such as pay-as-you-go, one-year reserve, or three-year reserve. The cost varies depending on the options selected and can be very expensive.

I give the solution a nine out of ten.

Depending on the organization's use case, there are multiple options available. For example, if the workload is medium, there are different products available such as Azure SQL Database for small or medium-sized data, and Azure Synapse Analytics for extremely large volumes of data. Ultimately, it is the use case of the organization that will determine the need for these particular services.

Our primary use case for Azure Synapse is as a data warehouse, for creation of data pipelines. It allows login into to one central workspace, manage our databases, and the entire warehouse. We can embed business intelligence (BI), using Power BI. This allows us to show visualizations all in one central place.

I think potential areas to improve on could be performance and if they offered a decoupled compute from storage kind of service that would be nice. But I don't think that is possible as it's a fundamental change in the underlying architecture and Microsoft won't make that decision easily. 

We have been using Microsoft Azure technology for the last 4-5 years, including Synapse.

As Synapse is hosted on the Azure cloud, it's very stable. 

The Azure Synapse service is highly scalable. 

The initial setup is very straight forward. Azure Synapse offers you one workspace where you can do everything, creation of your data warehouse, ETL pipelines using Azure Data Factory, Create storage and data marts. Also use Power BI for visualization. 

Before Synapse was available, all of these was offered as separate services and this is how a data warehouse was constructed. Synapse is one layer on top of this where we make use of one single workspace to initiate and manage the entire set of services that you need for creating and managing your data platform - Data Warehouses and marts using SQL Warehouse, ETL pipelines using Azure Data Factory, Data Lake using Azure Blob Storage, and it offers server-less SQL - meaning you can run queries without having to initiate an SQL database or SQL data warehouse instance. It also offers Spark compute to process non-structured data.

We are a Microsoft partner and have setup and built Azure Synapse based solutions for our manufacturing, energy and healthcare clients. We are very customer centric and build and manage solutions based on our clients needs. We recommend what the best technology stack is for them. 

If I hosted a Microsoft setup on premise, I would need to invest in licensing for different tools and services, SQL server, SSIS, SSRS, Power BI or SSAS. Compared to this if you use Azure Synapse, the return on investment is very high. You get rid of your hardware, licensing and you move to a subscription based pay as you use model. Your operational costs reduce and your optimization increases. Capital expenditure absolutely diminishes and you move to an OpEx model. 

Finally, the overall management of it is simplified as compared to on premise. This of course leads to high RoI. 

Azure Synapse is best for people who are already invested in Microsoft technologies, in particular those who already use Microsoft data warehousing services, including MS SQL-Server based data warehouse technology. For them, migrating to Azure is very straight forward and Synapse adoption stays easy. 

With Azure Synapse, there is no database installation, no licensing cost, no hardware setup, everything is available as a cloud service, you then pay for the service, pay only for what you use. 

With regards to pricing, as I said you pay for what you use. The amount of data you store and compute power contributes to your pricing. If I use Azure's blob storage, the pricing depends on how much I use. If I utilize Azure Data Factory, pricing depends on how much data I process through the ETL pipelines and so on.

For some customers we recommend Snowflake, for others Azure Synapse or Google BigQuery. In one of our cases we are building a solution with both Azure Synapse and Snowflake. 

We have approximately 20-25 team members with knowledge of Azure Synapse service capabilities. 

With regards to deployment and maintenance, an Azure Synapse based solution may need anywhere between 3-15 people. This depends on what type of warehouse and analytics you want to create, the number of reports and visualization. Typically a small team size would be of 3-4 people and a large team size would be of around 12-14 members.

We use this solution to create data pipelines and to improve the self-service environment for end users. We use all the functions of Microsoft Azure Synapse Analytics.

I like the keynotes and their simplicity. Like other Microsoft products, Microsoft Azure Synapse Analytics is simple to understand and use.

The linked services can be improved. We can create dynamic linked services to access a lot of databases but only those of the same type. For example, I can use the same linked services to access 11 SQL databases. However, if I have 11 SQL databases and five Oracle databases, I need two dynamic linked services. I cannot do it with only one linked service.

The UI also needs to be improved. When I have used Azure Synapse for programming with PySpark, Scala, or .NET, for example, the UI has been unstable. If I open two notebooks for programming, one notebook will queue the session of the other.

I've been using this solution for two years.

I would rate the stability at eight out of ten because Microsoft Azure Synapse Analytics is unstable during programming. 

It is simple to scale the service. I'd rate it at ten out of ten for scalability. We had 45 engineers who used the solution.

Technical support staff are knowledgeable. Whenever we have needed support, they have had a good process and have always had the answers we needed. I'd give technical support a rating of ten out of ten.

Positive

The initial setup is simple, and I would rate it at ten out of ten.

The pricing is competitive, but only when you pay upfront. If you pay as you go, it's not as competitive. I'd give pricing a rating of seven out of ten.

On a scale from one to ten, I would rate Microsoft Azure Synapse Analytics at eight.

Microsoft Azure Synapse Analytics is used for analytics.

The most valuable features of Microsoft Azure Synapse Analytics are the interface and the agility of the on-cloud platform.

Microsoft Azure Synapse Analytics's overall integration within the Azure ecosystem could improve. The native Microsoft solution versus another solution, such as Databricks, there are areas where there could be some improvements.

I have been using Microsoft Azure Synapse Analytics for approximately two years.

Microsoft Azure Synapse Analytics is stable.

The scalability of Microsoft Azure Synapse Analytics is very good.

We have hundreds of people using this solution.

Compared to the traditional data center approach, the cloud has pushed forward well for maintenance reduction. There is a 60 to 80 percent reduction.

My advice to others is this solution is not meant for small databases.

I rate Microsoft Azure Synapse Analytics an eight out of ten.

It's used for business case analysis and bandwidth utilization, etc. 

We are using its latest version.

It's feature-rich. It has a wide range of features. 

It's stable, but its stability could be better. However, we understand that it's in production, and new features are getting added and upgraded, so you do get hiccups sometimes.

I've been using this solution for more than four years.

It's quite stable. I'd rate it an eight out of ten in terms of stability. There are very few outages or errors during the configuration.

It's scalable. I'd rate it an eight out of ten in terms of scalability. Our clients are enterprises. 

I haven't contacted them, but it should be fairly good. I haven't had many issues with it.

I'd rate it a nine out of ten in terms of the setup. It has a GUI, which is awesome. Its deployment duration varies based on the case to case. We have quite a few SMEs for deployment. 

It doesn't require many people for maintenance.

It has been good so far. I'd recommend it to others. I'd rate it an eight out of ten.

We primarily use the solution for our cloud data warehouse. 

The usability has been excellent. 

It offers good integration. 

The product works well with other Azure products.

The initial setup is very convenient.

It is scalable.

We have found the solution to be stable. 

The security is fine.

We'd like the solution to have more analytics capability.

We would like more detail on reports so that they are easy to understand in terms of usage of data and services. It could be improved.

I've been using the solution for one year. 

It's a stable product. It is reliable. The performance is good. It doesn't crash or freeze. There aren't bugs or glitches. 

We can scale the product up and down according to our needs.

We have around 100 people using the solution right now. 

My customer uses it on a daily basis.

We are satisfied with technical support. I haven't used them too much.

Neutral

I've also used Snowflake. 

Azure Synapse is a more integrated platform. With Snowflake, you need to buy it separately and run it off Microsoft. Snowflake, however, has better cost control. Snowflake is comparatively cost-efficient.

It is a straightforward setup. It's easy and not overly complex. I have not measured how long it actually takes; however, I can say the implementation is convenient. My administrator handled the setup. I did not handle it directly myself. 

The solution can get expensive, and it's hard to monitor the costs. It does depend on how you utilize it. You can configure it in a certain way.

I'd rate the solution seven out of ten. 

I am the Associate Director of a company that is a partner of Synapse.

The solution offers strong scalability opportunities.

I would like to see more ready-to-use products from Synapse. Right now, everything seems a bit futuristic without much modern use.

The solution is scalable.

Price is dependent of the use case of the customer. Customers are not likely to only use this as an individual platform/database, they are likely to bundle it with the Azure stack.

Snowflake is a solution that is much more ahead of the game. There are many gaps in Synapse that Snowflake has fulfilled (i.e. usage, and storage).

Be sure to be clear about the ideal pipeline for Synapse, differentiate your needs, and validate based on use-case.

I would rate this solution a six out of ten overall.

The most valuable aspect of this Microsoft Azure Synapse Analytics is its consolidation of technical support from Microsoft, and its ability to securely host large quantities of data within the cloud environment. The overall ability to manage and maintain Big Data within the cloud provides a heightened level of efficiency, reliability, and support from Microsoft. This results in a superior user experience and an increased level of value for the end user.

There may be some challenges in terms of connecting with Virtual Networks (VNETs) to Microsoft Azure Synapse Analytics.

I have been using Microsoft Azure Synapse Analytics for approximately one year.

The stability of Microsoft Azure Synapse Analytics is great. It offers a high level of performance.

The scalability of Microsoft Azure Synapse Analytics is good.

The support from Microsoft Azure Synapse Analytics is excellent.

There is a steep learning curve to do the implementation.

I rate the initial setup of Microsoft Azure Synapse Analytics a seven out of ten.

The entire Microsoft Azure ecosystem is well connected and Microsoft has done well with the features.

My advice to others is to try out the solution, do your homework and research, and set up a POC environment.

I rate Microsoft Azure Synapse Analytics an eight out of ten.