DX NetOps

You might need to check the pre-reqs.

that might be causing

Regards
Sunny

Actually I am supporting this customer and I've been doing it for the past 14 years, since I initially installed Spectrum 8.1 in their environment.

Indeed, there are many flavors of devices and differences span from 4 ports access router to 10.000 ports Core infrastructure device. For my customer's needs, I'm confident that this will be working, as it is now with 4 SS.

On the other hand, I've seen a major service providers with close to  120 Core devices and a number of 450.000 interfaces in total. I do not know of a service provider to have 10.000 core devices with 10.00+ interfaces each, to be in this case, so 4SS should not suffice.

QoS monitoring in SPECTRUM is rather useless for just alarming purposes, as PM should be used better to track historically the queues and provide alarming the relevant values. Same goes for IPSLA, since the IPSLA reporting was removed from the CABI SRM reporting.

This tells me, that this customer, with 900+ devices cannot have all Core devices (10.000+ interfaces), as these figures that are presented would not suffice for such a situation.

The hardware assessment should be done based on gathered customer information and the figures that I saw here, I think they're grater then an actual implementation requirements.

What the system is good at and should be used extensively is topology relations for root cause analysis. That and Service Manager, but I haven't seen many customers using effectively Service Manager, as they find it rather hard to configure and maintain.

CA/Broadcom has always been conservative, as I've been saying with figures of number of supported devices. What happens if you run a performance assessment on the SS 6 months in its deployment and the load ends up to be 40%? It's a good 60% reserve for future usage? I've been there several times with the CA provided figures. 

Good Job then!
so we all need to make sure the customer has a supported environment that will meet their needs.

If your going to make a  significant investment in the software you don't what to put it on a minimal  h/w solution.

Himanshu: Your solution would be more than adequate for that environment and provide you with complete redundancy.

A bit more disc space on the SRM server .

Actually I am supporting this customer and I've been doing it for the past 14 years, since I initially installed Spectrum 8.1 in their environment.

Indeed, there are many flavors of devices and differences span from 4 ports access router to 10.000 ports Core infrastructure device. For my customer's needs, I'm confident that this will be working, as it is now with 4 SS.

On the other hand, I've seen a major service providers with close to  120 Core devices and a number of 450.000 interfaces in total. I do not know of a service provider to have 10.000 core devices with 10.00+ interfaces each, to be in this case, so 4SS should not suffice.

QoS monitoring in SPECTRUM is rather useless for just alarming purposes, as PM should be used better to track historically the queues and provide alarming the relevant values. Same goes for IPSLA, since the IPSLA reporting was removed from the CABI SRM reporting.

This tells me, that this customer, with 900+ devices cannot have all Core devices (10.000+ interfaces), as these figures that are presented would not suffice for such a situation.

The hardware assessment should be done based on gathered customer information and the figures that I saw here, I think they're grater then an actual implementation requirements.

What the system is good at and should be used extensively is topology relations for root cause analysis. That and Service Manager, but I haven't seen many customers using effectively Service Manager, as they find it rather hard to configure and maintain.

CA/Broadcom has always been conservative, as I've been saying with figures of number of supported devices. What happens if you run a performance assessment on the SS 6 months in its deployment and the load ends up to be 40%? It's a good 60% reserve for future usage? I've been there several times with the CA provided figures. 

Good Job then!
so we all need to make sure the customer has a supported environment that will meet their needs.

If your going to make a  significant investment in the software you don't what to put it on a minimal  h/w solution.

Himanshu: Your solution would be more than adequate for that environment and provide you with complete redundancy.

A bit more disc space on the SRM server .

Actually I am supporting this customer and I've been doing it for the past 14 years, since I initially installed Spectrum 8.1 in their environment.

Indeed, there are many flavors of devices and differences span from 4 ports access router to 10.000 ports Core infrastructure device. For my customer's needs, I'm confident that this will be working, as it is now with 4 SS.

On the other hand, I've seen a major service providers with close to  120 Core devices and a number of 450.000 interfaces in total. I do not know of a service provider to have 10.000 core devices with 10.00+ interfaces each, to be in this case, so 4SS should not suffice.

QoS monitoring in SPECTRUM is rather useless for just alarming purposes, as PM should be used better to track historically the queues and provide alarming the relevant values. Same goes for IPSLA, since the IPSLA reporting was removed from the CABI SRM reporting.

This tells me, that this customer, with 900+ devices cannot have all Core devices (10.000+ interfaces), as these figures that are presented would not suffice for such a situation.

The hardware assessment should be done based on gathered customer information and the figures that I saw here, I think they're grater then an actual implementation requirements.

What the system is good at and should be used extensively is topology relations for root cause analysis. That and Service Manager, but I haven't seen many customers using effectively Service Manager, as they find it rather hard to configure and maintain.

CA/Broadcom has always been conservative, as I've been saying with figures of number of supported devices. What happens if you run a performance assessment on the SS 6 months in its deployment and the load ends up to be 40%? It's a good 60% reserve for future usage? I've been there several times with the CA provided figures. 

Good Job then!
so we all need to make sure the customer has a supported environment that will meet their needs.

If your going to make a  significant investment in the software you don't what to put it on a minimal  h/w solution.

Himanshu: Your solution would be more than adequate for that environment and provide you with complete redundancy.

A bit more disc space on the SRM server .

Actually I am supporting this customer and I've been doing it for the past 14 years, since I initially installed Spectrum 8.1 in their environment.

Indeed, there are many flavors of devices and differences span from 4 ports access router to 10.000 ports Core infrastructure device. For my customer's needs, I'm confident that this will be working, as it is now with 4 SS.

On the other hand, I've seen a major service providers with close to  120 Core devices and a number of 450.000 interfaces in total. I do not know of a service provider to have 10.000 core devices with 10.00+ interfaces each, to be in this case, so 4SS should not suffice.

QoS monitoring in SPECTRUM is rather useless for just alarming purposes, as PM should be used better to track historically the queues and provide alarming the relevant values. Same goes for IPSLA, since the IPSLA reporting was removed from the CABI SRM reporting.

This tells me, that this customer, with 900+ devices cannot have all Core devices (10.000+ interfaces), as these figures that are presented would not suffice for such a situation.

The hardware assessment should be done based on gathered customer information and the figures that I saw here, I think they're grater then an actual implementation requirements.

What the system is good at and should be used extensively is topology relations for root cause analysis. That and Service Manager, but I haven't seen many customers using effectively Service Manager, as they find it rather hard to configure and maintain.

CA/Broadcom has always been conservative, as I've been saying with figures of number of supported devices. What happens if you run a performance assessment on the SS 6 months in its deployment and the load ends up to be 40%? It's a good 60% reserve for future usage? I've been there several times with the CA provided figures. 

Good Job then!
so we all need to make sure the customer has a supported environment that will meet their needs.

If your going to make a  significant investment in the software you don't what to put it on a minimal  h/w solution.

Himanshu: Your solution would be more than adequate for that environment and provide you with complete redundancy.

A bit more disc space on the SRM server .

Actually I am supporting this customer and I've been doing it for the past 14 years, since I initially installed Spectrum 8.1 in their environment.

Indeed, there are many flavors of devices and differences span from 4 ports access router to 10.000 ports Core infrastructure device. For my customer's needs, I'm confident that this will be working, as it is now with 4 SS.

On the other hand, I've seen a major service providers with close to  120 Core devices and a number of 450.000 interfaces in total. I do not know of a service provider to have 10.000 core devices with 10.00+ interfaces each, to be in this case, so 4SS should not suffice.

QoS monitoring in SPECTRUM is rather useless for just alarming purposes, as PM should be used better to track historically the queues and provide alarming the relevant values. Same goes for IPSLA, since the IPSLA reporting was removed from the CABI SRM reporting.

This tells me, that this customer, with 900+ devices cannot have all Core devices (10.000+ interfaces), as these figures that are presented would not suffice for such a situation.

The hardware assessment should be done based on gathered customer information and the figures that I saw here, I think they're grater then an actual implementation requirements.

What the system is good at and should be used extensively is topology relations for root cause analysis. That and Service Manager, but I haven't seen many customers using effectively Service Manager, as they find it rather hard to configure and maintain.

CA/Broadcom has always been conservative, as I've been saying with figures of number of supported devices. What happens if you run a performance assessment on the SS 6 months in its deployment and the load ends up to be 40%? It's a good 60% reserve for future usage? I've been there several times with the CA provided figures. 

Good Job then!
so we all need to make sure the customer has a supported environment that will meet their needs.

If your going to make a  significant investment in the software you don't what to put it on a minimal  h/w solution.

Himanshu: Your solution would be more than adequate for that environment and provide you with complete redundancy.

A bit more disc space on the SRM server .

Actually I am supporting this customer and I've been doing it for the past 14 years, since I initially installed Spectrum 8.1 in their environment.

Indeed, there are many flavors of devices and differences span from 4 ports access router to 10.000 ports Core infrastructure device. For my customer's needs, I'm confident that this will be working, as it is now with 4 SS.

On the other hand, I've seen a major service providers with close to  120 Core devices and a number of 450.000 interfaces in total. I do not know of a service provider to have 10.000 core devices with 10.00+ interfaces each, to be in this case, so 4SS should not suffice.

QoS monitoring in SPECTRUM is rather useless for just alarming purposes, as PM should be used better to track historically the queues and provide alarming the relevant values. Same goes for IPSLA, since the IPSLA reporting was removed from the CABI SRM reporting.

This tells me, that this customer, with 900+ devices cannot have all Core devices (10.000+ interfaces), as these figures that are presented would not suffice for such a situation.

The hardware assessment should be done based on gathered customer information and the figures that I saw here, I think they're grater then an actual implementation requirements.

What the system is good at and should be used extensively is topology relations for root cause analysis. That and Service Manager, but I haven't seen many customers using effectively Service Manager, as they find it rather hard to configure and maintain.

CA/Broadcom has always been conservative, as I've been saying with figures of number of supported devices. What happens if you run a performance assessment on the SS 6 months in its deployment and the load ends up to be 40%? It's a good 60% reserve for future usage? I've been there several times with the CA provided figures.