I have a load run with 150 concurrent users. however, after test execution I switched to virtual users activity chart, I saw virtual users with big id like 1205, with the id grows when test lasts longer. Shouldn't it be 1-150 or 0-149?
If you have the Percentage of new users greater than zero then you should expect the user id to continually increase throughout the test. The number of users will be 150, given the details in the question, but their identification numbers increase as each new user starts.
The effect of the user id can be seen in the Detail page of the load test results. Each horizontal line corresponds to one user id. When Percentage of new users is zero then you will see multiple tests executed for the same user. With 150 users there will be 150 horizontal lines each with multiple tests. When Percentage of new users is 100 then you will see a long diagonal running from bottom left corner to top right corner made of horizontal lines but with only one test on each horizontal line, ie only one test per user.
Related
I am looking to build a bot that typically requires two numbers with a different meaning (role) in the same utterance. Let's take the example of a StockMarket order assistent (fictional, as example)
Some example utterances:
Buy 100 MSFT stock at limit of 340
Get me 200 Apple at maximum 239.4
Buy 40 AMZN at market price
In LUIS portal, I have defined two entities
StockSymbol a List entity (for all stocks, linking their symbols and the names as synonyms).
number the prebuilt entity with two Roles: Amount and Limit
When specifying the utterances shown as example, it shows that the entities get recognized. But I cannot find a way to specify the roles for the different number entities in my sample utterances. (in the examples, the first number instance of number is the Amount, and if a second it there, that is typically the Limit role.
Anyone an idea on how to define this and set this up?
Best regards
There are 2 different ways to do this, First is to use roles for a prebuilt entity, go into the number prebuilt, click on Roles, add 2 different roles.
one for Amount another for Limit then you have to go in the utterances and label for the roles, you do that by going to the utterance, clicking on the # symbol on the right, selecting the number prebuilt, selecting the role, then highlighting the numbers with that role.
Second approach is to use ML entities, create 2 ML entities, one for Amount one for Limit. Add the number as a feature and make it a required feature, and then go ahead and label the Amounts with the Amount entity and the limits with the Limit entity directly.
I making a mobile app similar to facebook as a college project. Problem I am facing is say if I have 25 users and 24 users posted something on the site of which user1 made the first post. Now say the 25th user logs in after two days. Initially I was showing the feeds in FIFO order. But if I keep using FIFO, 25th user wont be able to see the last post unless he has undergone all the previous posts first. Same situation in LIFO. If we scale up the app, the problem statement would be huge as there would be some feeds which will never be seen. So how do I tackle this?
I thought of distributing users into groups and show the feeds within the groups first, and then to the neighbouring group and so on.
PS: I am using meteorjs for building the app.
So I have distributed the users into groups. I have attached a tag with every feed saying user from which group posted it.
Now I sort the feeds according to date-time at which they were created. Then I group them according to their user groups. Then I sort the feeds according to the no.of views (sorting within the group only).
Now I keep the sorted data as a list. To show the feeds to the user, I am trying to use trees. Say a user belonging to group 4 logs in. So I would use the group 4 data as node and create a binary tree. Then I would show all the feeds in the head first, then do a bfs traversal to show the feeds on the leaves.
Is there any other better method to make sure that every feed is seen atleast once by any user
So what I could think of is that you can divide the users into groups firstly. Then show the feeds from the same group. But show only new feeds here. Then show the feeds from neighbouring groups and move on. Then show the old feeds in the same manner.
I want to load test a website. The website is used mainly to submit picture albums and has two main worflows:
One workflow for user registration:
Create a new account
Enter username, password, email address
Agree to the terms
Upload an avatar picture
Another workflow to submit albums:
Login
Create a new album with name, location, etc.
Upload a picture
Tag picture
Repeat from 3 (0 to n times)
Submit
These workflows take quite some time to execute (submitting a new album takes up to 5 minute). If I want to generate 1 album submission per second, I would have 5 * 60 = 300 threads running.
Also there is constraint on the data:
A user cannot have two albums with the same name
A picture cannot be uploaded twice or by two different users
I am wondering how other websites with similar workflows have been load tested and what is the recommended approach.
Test Data is very important for any performance testing, the Test data used could within the application has to be unique or duplicate could work.
In your application it has to be Unique, so I recommend you need to create test data separately for each scenario with the sufficient number of records available to complete each transaction, without any interdependence.
For albums names, you can create the unique name to append the data & time within your script, for example - TData_0605_1205
Similarly, you need to build a good amount of Image repository that can last for 300 * total load test time(mins), the naming convention for these images can be incremental, say TImage1, TImage2, TImage3............TImagen , within your script, write a logic to pick the images in sequential fashion with incremental of 1, this number you can get from the thread count/iteration count depending on how you script your framework.
Hope this helps.
Thanks,
Bhaskar
Technical Head
Thought Frameworks Inc
C: +1-619-630-5034
I'm building an auction site using laravel 5 however I'm currently unsure of the best way to go around the bidding process.
I currently have it set-up so that once the user hits a bid button the script runs to place the bid, however if multiple users do this at the same time this causes issues with multiple bids with the same value. I thought about modifying this so that it would queue the bid that way only 1 bid is being processed at once however I believe there will be a better method.
If someone could point me in the correct direction it would be greatly appreciated.
One way to do it is to simply insert the bid with a high precision timestamp immediately, then check the table using a select and see if its actually a leading bid or not. The table should have an auto incrementing id, so even if two bids have the exact same timestamp, sorting also by id will tell you which one was actually received first.
I think I get the different between Users, Sessions, and Pageviews in Google Analytics.
A Visitor/ User can start several sessions even in the same day, of a website ... and each session is comprised of visiting many pages, possibly more than once.
Here's the confusion ... what the frack is the difference between Unique PageViews and Sessions, when applied against a certain page?
A Unique Pageview is basically --- was the page viewed at least ONCE during the session? If yes, +1, else, +0. Isn't that IDENTICAL to sessions?
At least, for an individual page.
I know you can visit 4 unique pages during a session, but a specific page should have the two things equal, right?
The report seems to be presenting the correct data.
Here is how I understand it : - For a given page, the Unique PageView counts the number of sessions that displayed this page one or more times. Unique Pageviews is the number of sessions during which the specified page was viewed at least once. A unique pageview is counted for each page URL + page Title combination.
But it is not how it shows in the case you have presented
Vs
In actual it shows Sessions and Unique Pageviews different.
Our explanation is that the unique pageviews are being shown in funnel which is same as Sessions but when applying filters based on session, you are being shown the session with Entrances on that page for some reason.
To confirm this – you can test adding a segment as shown below and match the output numbers.
1. When you check segment on a page, the session count is 5562 (as the report) I.e. all sessions in which that page was viewed.
When you check segment on a landing page, the session count is 600 (close to the filter report 589) I.e. all sessions in which that page was the landing page.
Ideally we also expect to see the number to be 5562, in the filtered view under the ‘sessions’ metric when we filter on a single page. Also, we expect the sessions to be same as unique pageviews but somehow the metric shown is referring to the ‘sessions that started with that page’ i.e. entrances.
Here is your actual answer, I think! :)
http://help.analyticsedge.com/googleanalytics/misunderstood-metrics-sessions-for-pages/
Basically, if you filter sessions results down to a page, you will see a count of how many sessions started on that page. (Sessions only get attached to the Entrance page).
Cheers.
The Unique Pageviews metric is an approximation of how many times the page has been viewed by a unique user in a given time period. By default, Google Analytics tracking cookie will remember a user on a device for 30 days, unless the user deletes cookies or uses a different device.
Sessions, on the other hand map to Entrances are is a metric that approximates how many times a given page-based dimension was the first hit of a session. A session is defined as an active period of browsing activity by a user of your site. After the session is initiated (someone comes to your site), the session timer will continue to run until either a)the user leaves your site, or b)there is a period of inactivity that lasts for 30 minutes. In either of those cases, the session will be closed. A new session would start when the user resumes activity after that 30 minute period of inactivity, or when they come back to your site.
This behaviour is expected behaviour.
The sessions count is the number of sessions who landed on that screen/page.
The unique page views are the number of sessions who viewed that page.
Eg.
Home screen: 100(sessions) 1000(unique page views)
Interpretation: 100 sessions landed on the home screen and 1000 sessions viewed the home screen( remaining 900 landed on some other screen but eventually viewed the home screen.
This is my understanding:
User A visits Page P1:
Page
Page views
Unique page views
User
Session
P1
1
1
1
1
Same session, user A visits page P2 from page 1 (P1):
Page
Page views
Unique page views
User
Session
P2
1
1
1
0
Notice the session count for page P2 does not increase (0) but unique page views for page P2 is 1.
I believe, because of cases like the above unique page views are greater than sessions count.
You can use this simulator for further understanding.