For me, the primary use case of Elasticsearch is log analysis, as it is a text-based search tool. To explain how it works, let's consider its role at the backend. Elasticsearch operates on keywords used to fetch data. This is in contrast to some databases, where operations might be based on a key order or a primary key, allowing for various maintenance and analysis tasks.

Many people use Elasticsearch to store their application logs in JSON format. These logs are indexed, facilitating efficient search and analysis. Additionally, Elasticsearch integrates well with tools like Grafana and Kibana, enabling users to create diverse dashboards for data visualization.

There's also the text-based search scenario. For instance, if a user wants to search for something using a specific keyword, Elasticsearch excels in this area by creating multiple indices.

Elasticsearch is a versatile tool that can store and retrieve information effectively, making it suitable for various applications across different industries.

Elasticsearch is a quick search engine tool. A good use case is saving metadata of your systems for data cataloging. Various systems, like those opened in metadata and similar applications, use Elasticsearch to store their text data. However, the major use case for many is to store application logs and build different dashboards on top of it.

The use of Elasticsearch is very specific. It is not helpful for storing your OLTP data. Elasticsearch's specific use is when you need to provide text-based search functionality. That's when Elasticsearch becomes relevant.

For instance, for log analysis or searching values, Elasticsearch performs very well. However, there are challenges with performance management and scalability, particularly how developers manage these aspects.

For example, Kubernetes is a popular choice as it offers the needed features to run your application and allows performance optimization in response to increased system load, and managing itself. If you plan to deploy Elasticsearch with limited or predefined resources, it may not be the ideal setup.

Therefore, it's better to create ultimate commerce capabilities for it. This is the challenge people are facing in the market and the solution for it. So, this answer combines two aspects: the challenge and its solution.

I have been using Elasticsearch for almost a year now. I'm comfortable working with it and understand its functionalities.

In our organization, it's not so much about the number of people as it is about the number of products utilizing it. Currently, we use Elasticsearch in more than 12 products.

It's become essential for any component that requires text-based functionality. Besides that, it's also used for logging to analyze application performance, peak times, etc. Elasticsearch is a basic component of the architecture for each of these products.

Most of our deployments are not exposed to the Internet or public networks; they're restricted to closed networks. We don’t frequently upgrade from previous versions unless a specific use case arises.

In such cases, we usually turn to the developer community for support.

Another scenario is when running the application in a careful mode, where the main requirement is to change the image name in the configuration. Then, we check for any changes or incompatibilities with previous versions. Upgrades can sometimes introduce issues if they’re not compatible with existing configuration files, but it's generally not too problematic to handle.

Deploying in Kubernetes is not complex. There are many resources in the market, like DevOps guys and guides, which make the process straightforward. The deployment can be done in a matter of minutes. You basically run a configuration file to set up your application, define replicas, and so on. It shouldn't take much time; even with an expert, it's a matter of a few hours.

However, the key lies in following best practices and configuring your files properly. If you follow the best practices, you'll likely face fewer issues. But if not, problems are inevitable.

It’s crucial to analyze these practices, considering factors like bandwidth, data volume, user interaction, and how it's read by different applications. These considerations help in managing resources and scalability, including scaling up and down your Elasticsearch container. These points are vital for running Elasticsearch efficiently, especially for text-based search applications.

You can deploy it as required. Elasticsearch is versatile; you can run it on Kubernetes, in the cloud, or on-premises. There is no limitation in terms of deployment options.

The cost varies based on factors like usage volume, network load, data storage size, and service utilization. If your usage isn't too extensive, the cost will be lower.

However, if you're dealing with high volumes, you'll need to reconsider the cost-effectiveness. If there are no challenges or bottlenecks in buying a service from a cloud service provider, that might be a viable option.

But if you're concerned about price or issues like exposing your data to the public cloud, then deploying on-premises and conducting stress testing becomes important. It’s a part of the learning and development process, not just a deployment for production.

You need to pass through testing processes in the development environment and then move to staging and production. This involves various tests to understand user access patterns, data push, and performance assessment. Deploying on your own requires considering all these factors. On the other hand, if you use a cloud service, many of these concerns aren't your responsibility.

If you're interested in using Elasticsearch as a search tool and for cloud data integration, comparing it with alternatives like Amazon Cloud Search or Azure Search is valid. Many cloud service providers that offer text-search services are utilizing Elasticsearch. They've implemented best practices and resolved a myriad of issues experienced by companies using Azure, AWS, or GCP.

These providers have integrated Elasticsearch into their cloud offerings effectively. Choosing their services might be preferable due to lower operational costs on your side.

In case of any disaster or issue, their development and DevOps teams are available to support you. However, if you face limitations, like client requirements prohibiting data storage in public or private clouds, then deploying Elasticsearch on-premises would be your alternative.

I would definitely rate it an eight out of ten, which is very good. The reason is the active community continuously working on it, and the support from contributors and the support team is notable. Because Elasticsearch is very specific in its use cases.

It excels in text-based search and creating dashboards for application logs. It provides results and functionality that are hard to find in alternative tools. So, if you have a use case that fits, Elasticsearch is a great service without any direct alternatives.

We use it for locating and retrieving documents, particularly in scenarios where the data lacks a predefined structure. These documents may encompass various types of information, such as logs or other records.

It is highly valuable because of its simplicity in maintenance, where most tasks are handled for you, and it offers a plethora of built-in features.

Currently, their focus seems to be on expanding integrations and introducing more external tools, somewhat diverging from enhancing the core product. While integrating with tools like agents for ingesting data from sources like firewalls is valuable, I believe prioritizing improvements to the core product would be more beneficial. For instance, the development of a multi-step query engine could significantly enhance user experience. The ability to execute queries, receive results, and then perform subsequent queries based on those results is a fundamental feature that, while achievable through code, seems to be lacking as a built-in capability. While they possess a robust infrastructure, the current upgrade process isn't seamless and can result in downtime. As a customer, this can be frustrating, especially when there are methods like replicating to a new instance, performing the upgrade, and then transitioning back, which could potentially minimize downtime. This is crucial in a cloud service where ensuring availability is paramount, considering the significant investment in such services.

I have been working with it for two years.

It offers good stability capabilities.

It is a scalable tool, but it's not impressive. The challenge arises when scaling out becomes prohibitively expensive. Instead of offering end-users the flexibility to specify the number of instances, there's a tendency to provide preconfigured packages. This approach may not be ideal, particularly for those seeking smaller scale-ups.

Their documentation is commendable as it provides a clear understanding of their offerings. Also, the accessibility to their support further enhances user-friendliness, making it a straightforward and user-friendly experience. While it may be slow, their competence in what they do is evident. I would rate it eight out of ten.

The initial setup was straightforward.

Setting up the system initially is quite straightforward, but when it comes to upgrades, the process becomes more challenging. It was an in-house deployment. The primary focus is on designing the solution, considering factors like the importance of replication, cluster size, speed, and disk space. I appreciate their approach of guiding you through these considerations, making it easier to grasp the bigger picture. This initial design phase is a complex but crucial step. Once that's sorted, the subsequent steps are relatively straightforward—just a few clicks to establish the baseline. If you're working on a standard deployment, it's a hassle-free process.

The pricing structure depends on the scalability steps. It begins as quite affordable and maintains affordability for a while. However, there's a turning point where it transitions from being reasonably priced to becoming notably expensive.

We've explored a few alternatives, but I believe Elasticsearch, particularly with Elastic and Elastic Cloud, stands out as the current industry standard. Opting for a widely used platform is advantageous due to the larger community it attracts. A substantial user base means more people to consult, numerous information sources, and a wealth of case studies. While there are smaller, medium, and even large alternatives, having around eighty percent of the community share provides a significant pool of expertise and resources to tap into.

The main reason we opted for it is because the installation is straightforward, and maintenance is made easy as they handle that aspect for you. The extensive knowledge base offers substantial assistance, making it less reliant on individual expertise. I believe it's a solid product, especially for beginners. While it's not free, it's well-suited for more complex tasks. Keep in mind that for intricate functionalities, you might need to develop and create tools beyond what Elastic Cloud offers. If you're considering a cloud-based solution for schema-less documents, Elasticsearch is a solid choice. On the other hand, if you have the resources to handle on-premises installation, I would recommend it for companies with the capability to manage the deployment themselves. Overall, I would rate it eight out of ten.

We save credentials, new account information, logs from Palantir Panorama, Firefox logs, traffic logs, GlobalProtect logs from our servers, and Active Directory new users. We're still improving this, but not very fast.

I appreciate that Elastic Enterprise Search is easy to use and that we have people on our team who are able to manage it effectively.

We are keeping an eye on other products like QRadar and Splunk in case they offer features that would benefit our company.

We currently use the free version of Elastic Search for some of our logs. However, if we were to use it more extensively, we would need to consider the pricing of the paid plans.

Another area of improvement is stability.

I have been using this solution for five years now.

I would rate the stability a seven out of ten. We faced a few issues.

I would rate the scalability a seven out of ten.

We don't use the support because we use the free version.

We were able to handle the deployment ourselves. We have one administrator and three users for this solution. So, there are four people in total.

I use the free version. We use the free version for some logs, but not extensive use.

Overall, I would rate the solution a seven out of ten. The free version is not very useful.