Skip to main content
74 votes

Optimal value for Nginx worker_connections

Let's take the pragmatic approach. All these limits are things that were hardcoded and designed in the past century when hardware was slow and expensive. We're in 2016 now, an average wall-mart ...
user5994461's user avatar
  • 3,028
29 votes

Optimal value for Nginx worker_connections

ulimit -a will tell you how many open files your system allows a process to use. Also, net.ipv4.ip_local_port_range sets the total range of sockets to enable per IP. So, your worker_connections ...
Marcel's user avatar
  • 1,865
13 votes
Accepted

Intel Xeon processors: Difference between 2S/4S and S2S/S4S scalability

The difference between S4S and 4S is the number of QPI-Links. 4S has 2 links per CPU, S4S and S8S have 3. On a 4S system CPUs form a ring, which means for every CPU there is another CPU with which it ...
Daniel Deppisch's user avatar
12 votes

Intel Xeon processors: Difference between 2S/4S and S2S/S4S scalability

Based on "6.5.8.4 MPSUP: Multiprocessor Support" in Intel® Xeon® Processor E7-4800/8800 v3 Product Families; Datasheet - Volume 1: Electrical, Mechanical and Thermal (pdf, page 139): This location ...
Ouroborus's user avatar
  • 221
5 votes

how to scale a high traffic server?

you are correct there is only so much Memory and only so many processors you can through at a server. Not to mention the fact that if you only have a single server running your application you have ...
Michael Brown's user avatar
3 votes
Accepted

How much data can SQL Server handle without a DBA / other professionals?

I don't think I can give you a magic "be worried here" number, where anything under that number is "okay" and anything over that number is "bad." That said, there are a number of red flags in your ...
Katherine Villyard's user avatar
2 votes
Accepted

Run shell script on the event of "possible SYN flooding"

In short: yes! But that depends a bit in the syslog daemon you're running. Syslog-ng allows that with the program() destination. Rsyslog offers actions.
HBruijn's user avatar
  • 80.8k
2 votes
Accepted

Handle "possible SYN flooding on port 443. Sending cookies."

Syn flood is kind of attack that is near to impossible to protect on single host. Check SynCookies cat /proc/sys/net/ipv4/tcp_syncookies and enable (set to 1) it if disabled. It help legit users ...
mmv-ru's user avatar
  • 714
2 votes

Optimal value for Nginx worker_connections

The appropriate sizing can be discovered through testing, as it is variable based on the type of traffic Nginx is handling. Theoretically, nginx can handle: max clients = worker_processes * ...
Sachin Gargya 's user avatar
2 votes
Accepted

AWS: Can we cosider having multiple EC2 instances and RDS for a single product as horizontal scaling?

Horizontal scaling typically means that you’ve got a fleet of undifferentiated servers where each server can serve any request. And you scale the capacity up and down by adding or removing more of the ...
MLu's user avatar
  • 25.6k
2 votes

Nosql autoscaling of compute

What if data is stored in a shared storage like CLVM which has less overhead ... This decouples the storage and compute problems of DB and can make compute horizontally scalable. ... Suggesting a ...
Rob's user avatar
  • 340
2 votes
Accepted

Intel Scalable Processors

The "scalable" part refers to the internal 2D mesh interconnection, which permit to efficiently connect a greater number of core than the previous ring topology. See here for more details.
shodanshok's user avatar
2 votes

Backup strategy for dedicated LEMP stack server

You are addressing redundancy. That is good. You could fail over to the back up server in an emergency. This is NOT A BACKUP solution. You want your backups, especially for a web app to go back in ...
Daniel Widrick's user avatar
1 vote
Accepted

What are the scalability strategies?

A scalability strategy is a plan with a course of action to follow when the usage pattern of your applications change. What can you, will you need to and wat are you going to do when your application ...
Bob's user avatar
  • 6,041
1 vote

What are the scalability concerns with pub/sub servers?

The actual memory usage of a single socket isn't that much. What does eat up memory is the state associated with which client is interested in which updates, and which client has already received a ...
Simon Richter's user avatar
1 vote

How (if at all) is KVM used with enterprise workloads?

KVM is widely used in the cloud, both private and public: OpenStack uses KVM AWS uses KVM Oracle OLVM uses KVM ... probably many more KVM is "just" a hypervisor with an API, many companies build ...
KeepLearning's user avatar
1 vote

Nosql autoscaling of compute

MongoDB, for instance, has the concept of the replica set for this situation. In this case multiple MongoDB instances serve the same data. If one fails, the others will continue serving the data. ...
Michael Hampton's user avatar
1 vote
Accepted

Servlet hosting with peak capacity

The rule of thumb would be to cache as much as you can. Preferably by one of the big CDN providers, like Akamai(expensive, but high quality), Cloudflare(free, but has it's issues), Fastly(I've no idea ...
Gothrek's user avatar
  • 551
1 vote

LoadBalancing for whole subnet

It's not possible and it doesn't make much sense to poll a whole network to see if a server exists or responds. You can use IPs or hostnames but they will be resolved at startup. You should add some ...
Federico Galli's user avatar
1 vote

Will EC2 server instance autoscale

No, it won't. An ec2 instance is just a virtual server. If you want to use autoscaling you have to set it up using the facilities AWS provides. AWS autoscaling scales horizontally, adding servers, ...
Tim's user avatar
  • 33k
1 vote

Will EC2 server instance autoscale

Not by default but it will if you set it up to autoscale - but you'll have to set that up as there's a cost implication and you need to tell it how you want it to work.
Chopper3's user avatar
  • 102k
1 vote

Optimal value for Nginx worker_connections

Marcel's answer really needs to be upvoted! If ulimits are set to a default value of around 1k, max_connections should be set around the same value otherwise there is no benefit to setting ...
Cryptophobia's user avatar
1 vote

Highly-available, Web-accessible and scalable deployment of statsd and graphite

There's a statsd proxy with consistent hashing, that makes is possible to spread statsd traffic between multiple statsd aggregators, each using its own set of metric names. It's a crucial scalability ...
DukeLion's user avatar
  • 3,299

Only top scored, non community-wiki answers of a minimum length are eligible